{"id":14248,"date":"2026-08-13T18:34:20","date_gmt":"2026-08-13T13:04:20","guid":{"rendered":"https:\/\/www.scaler.com\/blog\/?p=14248"},"modified":"2026-08-13T18:34:42","modified_gmt":"2026-08-13T13:04:42","slug":"nlp-project-ideas-with-source-code","status":"publish","type":"post","link":"https:\/\/www.scaler.com\/blog\/nlp-project-ideas-with-source-code\/","title":{"rendered":"NLP Project Ideas with Source Code: Text to LLMs"},"content":{"rendered":"\n

University NLP projects are already working with the tools behind today\u2019s language applications. Stanford\u2019s CS224N<\/a> course still covers the basics of NLP. Still, its 2026 coursework also includes BERT, pretraining, LoRA, and other parameter-efficient fine-tuning methods, prompting, RAG, tool use, and language-model evaluation. Students finish the course with a research project where they test their approach and report the results.<\/p>\n\n\n\n

For a student building nlp projects, there is a lot you can learn before getting into LLMs. Text classification, sentiment analysis, NER, and search can teach you how text is cleaned, represented, compared, and classified. Later projects can add embeddings and transformers, followed by document retrieval, RAG, structured extraction, and LLM evaluation.<\/p>\n\n\n\n

The 18 projects in this article cover these nlp projects with source code, starting with classical nlp project ideas and ending with LLM applications. You’ll also find the datasets, tools, and metrics used for the different projects.<\/p>\n\n\n\n

<\/span>NLP in 2026: Is It Still Useful<\/strong><\/span><\/h2>\n\n\n\n

If you’re searching for nlp projects in 2026, you may wonder whether techniques such as TF-IDF, stemming, and Naive Bayes are still worth learning when transformers and LLMs are now so common.<\/p>\n\n\n\n

And yes, they are still part of the technical picture. A TF-IDF classifier, for eg, can give you a baseline for a text-classification problem before you try a larger model. It can also be much cheaper to run when the task only requires a fixed prediction. An API call to GPT-5.6 Luna<\/a> currently costs $0.20 per million input tokens and $1.20 per million output tokens, so the difference becomes important when an application processes a large amount of text.<\/p>\n\n\n\n\nScaler Carousel<\/title>\n\n<!-- Plus Jakarta Sans Font -->\n<link href=\"https:\/\/fonts.googleapis.com\/css2?family=Plus+Jakarta+Sans:wght@400;500;600;700&display=swap\" rel=\"stylesheet\">\n<!-- Swiper CSS -->\n<link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.css\">\n\n<style>\n.scaler-carousel {\n font-family: 'Plus Jakarta Sans', sans-serif;\n max-width: 900px;\n margin: 24px auto;\n padding: 0;\n box-sizing: border-box;\n}\n.scaler-carousel *,\n.scaler-carousel *::before,\n.scaler-carousel *::after {\n box-sizing: border-box;\n}\n\n\/* Header *\/\n.scaler-header {\n text-align: center;\n margin-bottom: 20px;\n}\n.scaler-header-title {\n font-size: 22px;\n font-weight: 700;\n color: #1a1a2e;\n margin: 0 0 6px;\n line-height: 1.2;\n}\n.scaler-header-subtitle {\n font-size: 14px;\n color: #6b7280;\n margin: 0;\n font-weight: 500;\n}\n\n\/* Swiper *\/\n.swiper.scaler-swiper {\n padding-bottom: 48px;\n}\n.scaler-swiper .swiper-slide {\n height: auto;\n display: flex;\n align-items: stretch;\n}\n\/* Navigation arrows \u2014 styled to match the topics HorizontalScroll controls *\/\n.scaler-swiper .swiper-button-next,\n.scaler-swiper .swiper-button-prev {\n width: 46px;\n height: 46px;\n border-radius: 8px;\n background: #ffffff;\n box-shadow: 0 8px 16px 0 rgba(50, 71, 102, 0.16);\n top: calc(50% - 24px); \/* centre on the cards, ignoring the pagination space below *\/\n transform: translateY(-50%);\n margin-top: 0;\n}\n.scaler-swiper .swiper-button-prev { left: 10px; }\n.scaler-swiper .swiper-button-next { right: 10px; }\n\/* hide Swiper's default font-icon arrow and draw the topics chevron via a mask *\/\n.scaler-swiper .swiper-button-next::after,\n.scaler-swiper .swiper-button-prev::after {\n content: \"\";\n width: 20px;\n height: 20px;\n background-color: #61738e;\n -webkit-mask: url(\"data:image\/svg+xml,%3Csvg xmlns='http:\/\/www.w3.org\/2000\/svg' viewBox='0 0 32 32'%3E%3Cpath d='M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z'\/%3E%3C\/svg%3E\") no-repeat center \/ contain;\n mask: url(\"data:image\/svg+xml,%3Csvg xmlns='http:\/\/www.w3.org\/2000\/svg' viewBox='0 0 32 32'%3E%3Cpath d='M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z'\/%3E%3C\/svg%3E\") no-repeat center \/ contain;\n}\n.scaler-swiper .swiper-button-prev::after { transform: rotate(-90deg); }\n.scaler-swiper .swiper-button-next::after { transform: rotate(90deg); }\n.scaler-swiper .swiper-button-disabled {\n opacity: 0.35;\n pointer-events: none;\n}\n.scaler-swiper .swiper-pagination-bullet-active {\n background: #255aad;\n}\n\n\/* Card *\/\n.scaler-card {\n display: flex;\n flex-direction: column;\n gap: 14px;\n width: 100%;\n background: #ffffff;\n border: 1px solid #e5e7eb;\n border-radius: 16px;\n padding: 24px;\n box-shadow: 0 2px 8px rgba(0, 0, 0, 0.06);\n transition: box-shadow 0.2s ease;\n}\n.scaler-card:hover {\n box-shadow: 0 6px 16px rgba(0, 0, 0, 0.10);\n}\n\n\/* NSDC badge *\/\n.scaler-badge {\n display: inline-flex;\n align-items: center;\n align-self: flex-start;\n background: #f3f4f6;\n color: #111827;\n font-size: 12px;\n font-weight: 600;\n padding: 4px 10px;\n border-radius: 20px;\n white-space: nowrap;\n letter-spacing: 0.1px;\n}\n\n\/* Top: icon + title + description *\/\n.scaler-card-top {\n display: flex;\n align-items: flex-start;\n gap: 12px;\n min-height: 100px;\n}\n.scaler-icon-wrap {\n display: flex;\n align-items: center;\n justify-content: center;\n width: 64px;\n height: 64px;\n border-radius: 12px;\n background: #f3f4f6;\n color: #374151;\n flex-shrink: 0;\n}\n.scaler-icon-wrap svg {\n width: 26px;\n height: 26px;\n}\n.scaler-card-top-content {\n flex: 1;\n min-width: 0;\n display: flex;\n flex-direction: column;\n gap: 5px;\n}\n.scaler-title {\n font-size: 16px;\n font-weight: 700;\n color: #111827;\n margin: 0;\n line-height: 1.35;\n}\n.scaler-desc {\n font-size: 13px;\n color: #6b7280;\n margin: 0;\n line-height: 1.45;\n}\n\n\/* Stats row *\/\n.scaler-stats {\n display: flex;\n border-top: 1px solid #e5e7eb;\n border-bottom: 1px solid #e5e7eb;\n padding: 12px 0;\n}\n.scaler-stat {\n display: flex;\n flex-direction: column;\n align-items: center;\n gap: 2px;\n flex: 1;\n min-width: 0;\n text-align: center;\n}\n.scaler-stat:not(:last-child) {\n border-right: 1px solid #d1d5db;\n}\n.scaler-stat-value {\n font-size: 14px;\n font-weight: 700;\n color: #111827;\n white-space: nowrap;\n}\n.scaler-stat-label {\n font-size: 12px;\n color: #9ca3af;\n}\n\n\/* Company logos *\/\n.scaler-logos {\n display: flex;\n align-items: center;\n gap: 12px;\n flex-wrap: nowrap;\n}\n.scaler-company-logo {\n height: 20px !important;\n width: auto !important;\n max-width: 70px;\n object-fit: contain;\n display: inline-block;\n}\n.scaler-logo-more {\n font-size: 12px;\n color: #9ca3af;\n white-space: nowrap;\n}\n.scaler-logo-strip {\n display: block;\n width: 100% !important;\n max-width: 100%;\n min-width: 0;\n height: auto !important;\n object-fit: contain;\n}\n\n\/* CTA *\/\n.scaler-card-btns {\n margin-top: auto;\n}\n.scaler-card-btns button {\n display: flex;\n align-items: center;\n justify-content: center;\n gap: 8px;\n width: 100%;\n padding: 12px 16px;\n border: none;\n border-radius: 10px;\n background-color: #255aad;\n color: #ffffff;\n font-family: 'Plus Jakarta Sans', sans-serif;\n font-size: 14px;\n font-weight: 600;\n cursor: pointer;\n text-decoration: none;\n transition: opacity 0.15s ease;\n}\n.scaler-card-btns button:hover,\n.scaler-card-btns button:focus {\n opacity: 0.88;\n color: #ffffff;\n}\n.scaler-card-btns button svg {\n flex-shrink: 0;\n}\n\n@media (max-width: 1000px) {\n .scaler-carousel { max-width: 96vw; }\n}\n<\/style>\n\n\n<div class=\"scaler-carousel\">\n <div class=\"scaler-header\">\n <h2 class=\"scaler-header-title\"><span class=\"ez-toc-section\" id=\"transform-your-career\"><\/span>Transform Your Career<span class=\"ez-toc-section-end\"><\/span><\/h2>\n <p class=\"scaler-header-subtitle\">Choose from our industry-leading programs designed for career success<\/p>\n <\/div>\n\n <div class=\"swiper scaler-swiper\">\n <div class=\"swiper-wrapper\">\n\n <!-- CARD 1 -->\n <div class=\"swiper-slide\">\n <div class=\"scaler-card\">\n <div class=\"scaler-badge\">NSDC Certified<\/div>\n <div class=\"scaler-card-top\">\n <div class=\"scaler-icon-wrap\">\n <svg width=\"26\" height=\"26\" viewBox=\"0 0 24 24\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><polyline points=\"16 18 22 12 16 6\"\/><polyline points=\"8 6 2 12 8 18\"\/><\/svg>\n <\/div>\n <div class=\"scaler-card-top-content\">\n <h3 class=\"scaler-title\">Modern Software and AI Engineering Program<\/h3>\n <p class=\"scaler-desc\">Master full-stack development with AI integration<\/p>\n <\/div>\n <\/div>\n <div class=\"scaler-stats\">\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">12 Months<\/span><span class=\"scaler-stat-label\">Duration<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">AI-Led<\/span><span class=\"scaler-stat-label\">Curriculum<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">Career Support<\/span><span class=\"scaler-stat-label\">Support<\/span><\/div>\n <\/div>\n <div class=\"scaler-logos\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/2\/2f\/Google_2015_logo.svg\" alt=\"Google\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/a\/a9\/Amazon_logo.svg\" alt=\"Amazon\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/4\/42\/Paytm_logo.png\" alt=\"Paytm\" loading=\"lazy\">\n <span class=\"scaler-logo-more\">+1000 more<\/span>\n <\/div>\n <div class=\"scaler-card-btns\">\n <button onclick=\"window.open('https:\/\/www.scaler.com\/academy\/', '_blank')\">\n Go to Program\n <svg width=\"14\" height=\"14\" viewBox=\"0 0 16 16\" fill=\"none\" aria-hidden=\"true\"><path d=\"M3 8H13M13 8L9 4M13 8L9 12\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <!-- CARD 2 -->\n <div class=\"swiper-slide\">\n <div class=\"scaler-card\">\n <div class=\"scaler-badge\">NSDC Certified<\/div>\n <div class=\"scaler-card-top\">\n <div class=\"scaler-icon-wrap\">\n <svg width=\"26\" height=\"26\" viewBox=\"0 0 24 24\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><ellipse cx=\"12\" cy=\"5\" rx=\"9\" ry=\"3\"\/><path d=\"M3 5v14c0 1.66 4.03 3 9 3s9-1.34 9-3V5\"\/><path d=\"M3 12c0 1.66 4.03 3 9 3s9-1.34 9-3\"\/><\/svg>\n <\/div>\n <div class=\"scaler-card-top-content\">\n <h3 class=\"scaler-title\">Modern Data Science and ML with specialisation in AI<\/h3>\n <p class=\"scaler-desc\">Advanced data science techniques with AI specialization<\/p>\n <\/div>\n <\/div>\n <div class=\"scaler-stats\">\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">12 Months<\/span><span class=\"scaler-stat-label\">Duration<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">AI-Led<\/span><span class=\"scaler-stat-label\">Curriculum<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">Career Support<\/span><span class=\"scaler-stat-label\">Support<\/span><\/div>\n <\/div>\n <div class=\"scaler-logos\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/2\/2f\/Google_2015_logo.svg\" alt=\"Google\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/a\/a9\/Amazon_logo.svg\" alt=\"Amazon\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/4\/42\/Paytm_logo.png\" alt=\"Paytm\" loading=\"lazy\">\n <span class=\"scaler-logo-more\">+1000 more<\/span>\n <\/div>\n <div class=\"scaler-card-btns\">\n <button onclick=\"window.open('https:\/\/www.scaler.com\/data-science-course\/', '_blank')\">\n Go to Program\n <svg width=\"14\" height=\"14\" viewBox=\"0 0 16 16\" fill=\"none\" aria-hidden=\"true\"><path d=\"M3 8H13M13 8L9 4M13 8L9 12\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <!-- CARD 3 -->\n <div class=\"swiper-slide\">\n <div class=\"scaler-card\">\n <div class=\"scaler-badge\">NSDC Certified<\/div>\n <div class=\"scaler-card-top\">\n <div class=\"scaler-icon-wrap\">\n <svg width=\"26\" height=\"26\" viewBox=\"0 0 24 24\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M12 2L13.5 8.5L20 10L13.5 11.5L12 18L10.5 11.5L4 10L10.5 8.5L12 2Z\"\/><\/svg>\n <\/div>\n <div class=\"scaler-card-top-content\">\n <h3 class=\"scaler-title\">Advanced AIML with Specialisation in Agentic AI<\/h3>\n <p class=\"scaler-desc\">Deep dive into AIML with focus on Agentic systems<\/p>\n <\/div>\n <\/div>\n <div class=\"scaler-stats\">\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">12 Months<\/span><span class=\"scaler-stat-label\">Duration<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">AI-Led<\/span><span class=\"scaler-stat-label\">Curriculum<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">Career Support<\/span><span class=\"scaler-stat-label\">Support<\/span><\/div>\n <\/div>\n <div class=\"scaler-logos\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/2\/2f\/Google_2015_logo.svg\" alt=\"Google\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/a\/a9\/Amazon_logo.svg\" alt=\"Amazon\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/4\/42\/Paytm_logo.png\" alt=\"Paytm\" loading=\"lazy\">\n <span class=\"scaler-logo-more\">+1000 more<\/span>\n <\/div>\n <div class=\"scaler-card-btns\">\n <button onclick=\"window.open('https:\/\/www.scaler.com\/ai-machine-learning-course\/', '_blank')\">\n Go to Program\n <svg width=\"14\" height=\"14\" viewBox=\"0 0 16 16\" fill=\"none\" aria-hidden=\"true\"><path d=\"M3 8H13M13 8L9 4M13 8L9 12\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <!-- CARD 4 -->\n <div class=\"swiper-slide\">\n <div class=\"scaler-card\">\n <div class=\"scaler-badge\">NSDC Certified<\/div>\n <div class=\"scaler-card-top\">\n <div class=\"scaler-icon-wrap\">\n <svg width=\"26\" height=\"26\" viewBox=\"0 0 24 24\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\">\n <path d=\"M17.5 19H9a7 7 0 1 1 6.71-9h1.79a4.5 4.5 0 1 1 0 9Z\" \/>\n <path d=\"M9.2 15.2a3 3 0 1 0 4.24-4.24\" \/>\n <\/svg>\n <\/div>\n <div class=\"scaler-card-top-content\">\n <h3 class=\"scaler-title\">DevOps, Cloud & AI Platform Engineering<\/h3>\n <p class=\"scaler-desc\">Build and manage AI-powered cloud infrastructure<\/p>\n <\/div>\n <\/div>\n <div class=\"scaler-stats\">\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">12 Months<\/span><span class=\"scaler-stat-label\">Duration<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">AI-Led<\/span><span class=\"scaler-stat-label\">Curriculum<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">Career Support<\/span><span class=\"scaler-stat-label\">Support<\/span><\/div>\n <\/div>\n <div class=\"scaler-logos\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/2\/2f\/Google_2015_logo.svg\" alt=\"Google\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/a\/a9\/Amazon_logo.svg\" alt=\"Amazon\" loading=\"lazy\">\n <img decoding=\"async\" class=\"scaler-company-logo\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/4\/42\/Paytm_logo.png\" alt=\"Paytm\" loading=\"lazy\">\n <span class=\"scaler-logo-more\">+1000 more<\/span>\n <\/div>\n <div class=\"scaler-card-btns\">\n <button onclick=\"window.open('https:\/\/www.scaler.com\/devops-course\/', '_blank')\">\n Go to Program\n <svg width=\"14\" height=\"14\" viewBox=\"0 0 16 16\" fill=\"none\" aria-hidden=\"true\"><path d=\"M3 8H13M13 8L9 4M13 8L9 12\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <!-- CARD 5 -->\n <div class=\"swiper-slide\">\n <div class=\"scaler-card\">\n <div class=\"scaler-badge\">NSDC Certified<\/div>\n <div class=\"scaler-card-top\">\n <div class=\"scaler-icon-wrap\">\n <svg width=\"26\" height=\"26\" viewBox=\"0 0 24 24\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\">\n <line x1=\"3\" y1=\"22\" x2=\"21\" y2=\"22\" \/>\n <line x1=\"6\" y1=\"18\" x2=\"6\" y2=\"11\" \/>\n <line x1=\"10\" y1=\"18\" x2=\"10\" y2=\"11\" \/>\n <line x1=\"14\" y1=\"18\" x2=\"14\" y2=\"11\" \/>\n <line x1=\"18\" y1=\"18\" x2=\"18\" y2=\"11\" \/>\n <polygon points=\"12 2 20 7 4 7\" \/>\n <\/svg>\n <\/div>\n <div class=\"scaler-card-top-content\">\n <h3 class=\"scaler-title\">AI Engineering Advanced Certification by IIT-Roorkee<\/h3>\n <p class=\"scaler-desc\">Premier AI engineering certification from IIT-Roorkee<\/p>\n <\/div>\n <\/div>\n <div class=\"scaler-stats\">\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">3 Months<\/span><span class=\"scaler-stat-label\">Duration<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">AI-Led<\/span><span class=\"scaler-stat-label\">Curriculum<\/span><\/div>\n <div class=\"scaler-stat\"><span class=\"scaler-stat-value\">Career Support<\/span><span class=\"scaler-stat-label\">Support<\/span><\/div>\n <\/div>\n <div class=\"scaler-logos\">\n <img decoding=\"async\" class=\"scaler-logo-strip\" src=\"https:\/\/scaler-blog-prod-wp-content.s3.ap-south-1.amazonaws.com\/wp-content\/uploads\/2026\/07\/08135139\/iitr-icons-strip.png\" alt=\"Program highlights\" loading=\"lazy\">\n <\/div>\n <div class=\"scaler-card-btns\">\n <button onclick=\"window.open('https:\/\/www.scaler.com\/iit-roorkee-advanced-ai-engineering-course\/', '_blank')\">\n Go to Program\n <svg width=\"14\" height=\"14\" viewBox=\"0 0 16 16\" fill=\"none\" aria-hidden=\"true\"><path d=\"M3 8H13M13 8L9 4M13 8L9 12\" stroke=\"currentColor\" stroke-width=\"2\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <!-- Continue for remaining cards (6\u20138) using same pattern -->\n\n <\/div>\n <div class=\"swiper-button-next\"><\/div>\n <div class=\"swiper-button-prev\"><\/div>\n <div class=\"swiper-pagination\"><\/div>\n <\/div>\n<\/div>\n\n<!-- Swiper.js -->\n<script src=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.js\"><\/script>\n<script>\ndocument.addEventListener(\"DOMContentLoaded\", function () {\n const pathParts = window.location.pathname.split(\"\/\").filter(Boolean);\n const slug = pathParts[pathParts.length - 1] || \"\";\n const readableSlug = encodeURIComponent(slug);\n const buttons = document.querySelectorAll(\".scaler-card-btns button\");\n\n buttons.forEach((btn) => {\n const baseURL = btn.getAttribute(\"onclick\").match(\/'(.*?)'\/)[1];\n if (!baseURL.includes(\"utm_source=\")) {\n const separator = baseURL.includes(\"?\") ? \"&\" : \"?\";\n const updatedURL = `${baseURL}${separator}utm_source=blog&utm_medium=program_carousel&utm_content=${readableSlug}`;\n btn.setAttribute(\"onclick\", `window.open('${updatedURL}', '_blank')`);\n }\n });\n\n new Swiper(\".scaler-swiper\", {\n slidesPerView: 1,\n spaceBetween: 28,\n grabCursor: true,\n navigation: { nextEl: \".swiper-button-next\", prevEl: \".swiper-button-prev\" },\n pagination: { el: \".swiper-pagination\", dynamicBullets: true, clickable: true },\n breakpoints: { 1000: { slidesPerView: 2 } },\n keyboard: { enabled: true },\n });\n});\n<\/script>\n\n\n\n<p class=\"wp-block-paragraph\">You may also encounter these concepts when applying for NLP roles. TF-IDF, tokenisation, Naive Bayes, embeddings, and evaluation metrics such as Precision, Recall, and F1-score are still common parts of NLP interviews. Knowing them also makes it easier to understand why a particular model was chosen instead of simply reaching for an LLM.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This is why a 2026 nlp projects portfolio does not need to discard older methods. A project using a traditional classifier can show how you approach a problem, establish a baseline, and evaluate a model just as a transformer or LLM project can demonstrate newer skills.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">If you’re not sure what to learn after these projects, this <a href=\"https:\/\/www.scaler.com\/blog\/nlp-roadmap\/\">NLP Roadmap<\/a> can help you plan your next steps.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"era-1-classical-nlp-baselines-projects-1-6\"><\/span><strong>Era 1: Classical NLP Baselines (Projects 1-6)<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Classical NLP uses statistical methods and text features to work with language. Techniques such as TF-IDF, Naive Bayes, and NER can be used to classify text, find important information, or retrieve relevant documents without relying on a large language model. These methods are still useful when a task needs a fast, lightweight model or when you want to understand exactly how the system reaches its output.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The six nlp project ideas in this section cover TF-IDF, text classification, sentiment analysis, keyword extraction, Named Entity Recognition (NER), and information retrieval. For the implementation, you’ll work with libraries such as <a href=\"https:\/\/scikit-learn.org\/stable\/index.html\" target=\"_blank\" rel=\"noopener\">scikit-learn<\/a> and <a href=\"https:\/\/spacy.io\/models\" target=\"_blank\" rel=\"noopener\">spaCy<\/a>, using them for different parts of the NLP workflow. The projects take you through classification, sentiment analysis, information extraction, and search before you move on to transformer models and LLM applications.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>1. Spam Email Classifier using TF-IDF and Scikit-learn<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Email providers process millions of messages every day, making spam detection one of the earliest and most practical NLP applications. In this text classification project, train a machine-learning model to distinguish spam from legitimate emails. Start with a labelled email dataset, clean the text, and use TF-IDF to turn each email into numerical features. With TfidfVectorizer from scikit-learn, train a Multinomial Naive Bayes or Logistic Regression classifier on these features.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Test the trained model on emails it has not seen before and check accuracy, precision, recall, and F1-score. Precision is particularly important because a false positive could send a legitimate email to the spam folder.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, pandas, scikit-learn, and a labelled spam-email dataset.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.scaler.com\/topics\/nlp\/nlp-text-classification-using-scikit-learn\/\">Text Classification Using Scikit-learn<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> TF-IDF, text preprocessing, feature extraction, Naive Bayes, Logistic Regression, text classification, model evaluation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>2. Twitter sentiment analysis project<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Social media gives businesses instant feedback on product launches, marketing campaigns, and customer experiences, but going through thousands of posts manually isn’t quite practical. Sentiment analysis can sort these posts into categories such as positive, negative, or neutral.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Use a labelled collection of tweets and train a classifier to predict the sentiment of new posts. Tweets are short and informal, so clean the text carefully. Handle hashtags, emojis, mentions, links, abbreviations, and repeated characters without removing useful information. You can then represent the tweets with TF-IDF or word embeddings and train a classification model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Compare different preprocessing choices and check precision, recall, and F1-score for each sentiment class. This will especially come of use when one sentiment appears much more often than the others.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, pandas, scikit-learn, a labelled Twitter sentiment dataset, and a text-preprocessing setup.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.scaler.com\/topics\/data-science\/social-media-sentiment-analysis\/\">Social Media Sentiment Analysis<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Text preprocessing, sentiment classification, noisy-text handling, TF-IDF, word embeddings, model evaluation.<\/p>\n\n\n<meta charset=\"UTF-8\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n\n <!-- Source Sans Pro (topics font) -->\n <link href=\"https:\/\/fonts.googleapis.com\/css2?family=Source+Sans+3:wght@400;600;700&display=swap\" rel=\"stylesheet\">\n <link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.css\">\n <script src=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.js\"><\/script>\n\n <style>\n :root {\n \/* topics design tokens (variables.scss) *\/\n --scaler-primary: #005bb3; \/* primary-color-80 *\/\n --scaler-primary-30: #8acaff; \/* primary-color-30 *\/\n --scaler-brand-blue: #255aad;\n --scaler-live: #ee001d; \/* accent-v5-60 *\/\n --scaler-live-bg: #ffe6e9; \/* accent-v5-10 *\/\n --scaler-card-bg: #ffffff; \/* neutral-scale-0 *\/\n --scaler-surface: #f4f6f9; \/* neutral-scale-2 *\/\n --scaler-text-primary: #1a2c47; \/* neutral-scale-12 *\/\n --scaler-text-secondary: #61738e; \/* neutral-scale-9 *\/\n --scaler-border: #d6dee5; \/* neutral-scale-4 *\/\n --scaler-radius: 16px;\n --scaler-radius-sm: 8px;\n --scaler-shadow: 0 8px 16px 0 rgba(50, 71, 102, 0.16);\n }\n\n .scaler-events-carousel * { box-sizing: border-box; }\n\n .scaler-events-carousel {\n font-family: \"Source Sans 3\", \"Source Sans Pro\", \"Roboto\", Arial, sans-serif;\n padding: 24px;\n position: relative;\n width: 100%;\n }\n .scaler-events-carousel__inner {\n max-width: 1000px;\n margin: 0 auto;\n position: relative;\n width: 100%;\n }\n\n \/* Header \u2014 calendar icon + \"Live Masterclasses\" *\/\n .scaler-events-header {\n display: flex;\n align-items: center;\n gap: 8px;\n margin-bottom: 8px;\n padding: 8px 0;\n }\n .scaler-events-header__icon { display: inline-flex; color: var(--scaler-text-primary); }\n .scaler-events-header__icon svg { width: 24px; height: 24px; }\n .scaler-events-header__title {\n font-size: 18px;\n font-weight: 700;\n color: var(--scaler-text-primary);\n line-height: 1.3;\n margin: 0;\n }\n\n \/* Swiper wrap (holds side arrows) *\/\n .scaler-events-swiper-wrap { position: relative; }\n .scaler-events-carousel .swiper { width: 100%; padding: 4px; }\n .scaler-event-swiper .swiper-slide {\n width: 296px; \/* topics card_container width *\/\n height: auto;\n display: flex;\n }\n\n \/* Event Card \u2014 matches topics EventCard *\/\n .scaler-event-card {\n background: var(--scaler-card-bg);\n border: 1px solid var(--scaler-border);\n border-radius: var(--scaler-radius);\n padding: 16px;\n display: flex;\n flex-direction: column;\n height: 100%;\n width: 100%;\n cursor: pointer;\n }\n\n \/* Header row: headshot + title\/name *\/\n .scaler-event-card__header { display: flex; align-items: flex-start; }\n .scaler-event-card__image-wrapper {\n position: relative;\n width: 88px;\n height: 88px;\n min-width: 88px;\n border-radius: var(--scaler-radius-sm);\n overflow: hidden;\n background: var(--scaler-primary-30);\n }\n .scaler-event-card__image {\n position: absolute;\n top: 0; left: 0;\n width: 100%; height: 100%;\n object-fit: cover;\n }\n \/* attendee-count footer (default) *\/\n .scaler-event-card__img-footer {\n position: absolute;\n left: 0; right: 0; bottom: 0;\n display: flex;\n align-items: center;\n justify-content: center;\n gap: 4px;\n padding: 1px 8px;\n background: var(--scaler-surface);\n color: var(--scaler-text-primary);\n font-size: 13px;\n line-height: 22px;\n }\n .scaler-event-card__img-footer svg { width: 14px; height: 14px; }\n \/* live chip (shown when status === current) *\/\n .scaler-event-card__live-badge {\n position: absolute;\n left: 0; right: 0; bottom: 0;\n display: flex;\n align-items: center;\n justify-content: center;\n gap: 5px;\n padding: 1px 8px;\n background: var(--scaler-live-bg);\n color: var(--scaler-live);\n font-size: 12px;\n line-height: 22px;\n text-transform: uppercase;\n letter-spacing: 0.18em;\n }\n .scaler-event-card__live-badge::before {\n content: \"\";\n width: 6px; height: 6px;\n background: var(--scaler-live);\n border-radius: 50%;\n }\n .scaler-event-card__text { margin-left: 16px; min-width: 0; }\n .scaler-event-card__title {\n font-size: 20px;\n font-weight: 600;\n color: var(--scaler-text-primary);\n margin: 0;\n line-height: 1.4;\n display: -webkit-box;\n -webkit-line-clamp: 2;\n -webkit-box-orient: vertical;\n overflow: hidden;\n }\n .scaler-event-card__speaker {\n margin-top: 8px;\n font-size: 14px;\n line-height: 1.4;\n color: var(--scaler-text-primary);\n }\n\n \/* Takeaways *\/\n .scaler-event-card__takeaways {\n margin-top: 12px;\n min-height: 50px;\n max-height: 50px;\n overflow: hidden;\n }\n .scaler-event-card__takeaway-item {\n display: flex;\n align-items: flex-start;\n gap: 6px;\n }\n .scaler-event-card__dot {\n width: 4px; height: 4px;\n min-width: 4px;\n border-radius: 50%;\n background: var(--scaler-text-secondary);\n margin-top: 11px;\n }\n .scaler-event-card__takeaway-text {\n flex: 1;\n min-width: 0;\n font-size: 14px;\n line-height: 24px;\n color: var(--scaler-text-primary);\n white-space: nowrap;\n overflow: hidden;\n text-overflow: ellipsis;\n }\n .scaler-event-card__knowmore,\n .scaler-event-card__knowmore:link,\n .scaler-event-card__knowmore:visited {\n flex-shrink: 0;\n font-size: 14px;\n line-height: 24px;\n color: var(--scaler-primary) !important;\n text-decoration: none;\n white-space: nowrap;\n }\n .scaler-event-card__knowmore:hover { text-decoration: underline; }\n\n \/* Date + duration *\/\n .scaler-event-card__meta {\n display: flex;\n align-items: center;\n justify-content: space-between;\n margin-top: 20px;\n font-size: 13px;\n line-height: 24px;\n color: var(--scaler-text-primary);\n }\n .scaler-event-card__duration {\n display: flex;\n align-items: center;\n gap: 4px;\n }\n .scaler-event-card__duration svg { width: 15px; height: 15px; }\n\n .scaler-event-card__spacer { flex-grow: 1; min-height: 4px; }\n\n \/* CTA \u2014 outlined \"Register with 1-Click\" *\/\n .scaler-event-card__cta,\n .scaler-event-card__cta:link,\n .scaler-event-card__cta:visited {\n display: flex;\n align-items: center;\n justify-content: center;\n width: 100%;\n margin-top: 16px;\n padding: 8px;\n background: transparent;\n color: var(--scaler-primary) !important;\n font-family: inherit;\n font-size: 14px;\n font-weight: 600;\n line-height: 24px;\n text-decoration: none;\n border: 1px solid var(--scaler-primary);\n border-radius: var(--scaler-radius-sm);\n cursor: pointer;\n transition: background 0.2s ease, color 0.2s ease;\n }\n .scaler-event-card__cta:hover,\n .scaler-event-card__cta:focus {\n background: var(--scaler-primary);\n color: #fff !important;\n }\n\n \/* Side navigation arrows *\/\n .scaler-events-nav__btn {\n position: absolute !important;\n top: 50% !important;\n transform: translateY(-50%) !important;\n z-index: 2;\n width: 46px !important;\n height: 46px !important;\n min-width: 46px;\n display: flex !important;\n align-items: center;\n justify-content: center;\n padding: 0 !important;\n background: var(--scaler-card-bg) !important;\n border: none !important;\n border-radius: var(--scaler-radius-sm) !important;\n box-shadow: var(--scaler-shadow) !important;\n color: var(--scaler-text-secondary) !important;\n cursor: pointer;\n transition: opacity 0.2s ease, visibility 0.2s ease;\n }\n .scaler-nav-prev { left: 8px !important; }\n .scaler-nav-next { right: 8px !important; }\n .scaler-events-nav__btn svg {\n width: 20px !important;\n height: 20px !important;\n fill: var(--scaler-text-secondary) !important;\n }\n .scaler-events-nav__btn.swiper-button-disabled {\n opacity: 0;\n visibility: hidden;\n pointer-events: none;\n }\n\n \/* View All Events *\/\n .scaler-events-viewall,\n .scaler-events-viewall:link,\n .scaler-events-viewall:visited {\n display: flex;\n align-items: center;\n justify-content: center;\n gap: 6px;\n margin-top: 12px;\n font-size: 14px;\n font-weight: 600;\n color: var(--scaler-brand-blue) !important;\n text-decoration: none;\n }\n .scaler-events-viewall:hover { text-decoration: underline; }\n .scaler-events-viewall svg { width: 16px; height: 16px; fill: var(--scaler-brand-blue); }\n\n \/* hide swiper defaults *\/\n .scaler-events-pagination { display: none; }\n .scaler-events-carousel .swiper-button-prev,\n .scaler-events-carousel .swiper-button-next { display: none; }\n\n \/* Skeleton & Empty States *\/\n .scaler-event-card--skeleton { pointer-events: none; }\n .scaler-event-card--skeleton .scaler-event-card__image-wrapper,\n .scaler-event-card--skeleton .scaler-event-card__title,\n .scaler-event-card--skeleton .scaler-event-card__speaker,\n .scaler-event-card--skeleton .scaler-event-card__takeaways,\n .scaler-event-card--skeleton .scaler-event-card__meta,\n .scaler-event-card--skeleton .scaler-event-card__cta {\n background: linear-gradient(90deg, var(--scaler-border) 25%, var(--scaler-surface) 50%, var(--scaler-border) 75%);\n background-size: 200% 100%;\n animation: shimmer 1.5s infinite;\n color: transparent !important;\n border-color: transparent !important;\n border-radius: 4px;\n }\n .scaler-event-card--skeleton .scaler-event-card__title { min-height: 44px; }\n .scaler-event-card--skeleton .scaler-event-card__speaker { min-height: 16px; }\n .scaler-event-card--skeleton .scaler-event-card__meta { min-height: 24px; }\n .scaler-event-card--skeleton .scaler-event-card__cta { min-height: 40px; }\n .scaler-event-card--skeleton .scaler-event-card__image,\n .scaler-event-card--skeleton .scaler-event-card__img-footer,\n .scaler-event-card--skeleton .scaler-event-card__live-badge { display: none; }\n\n @keyframes shimmer {\n 0% { background-position: 200% 0; }\n 100% { background-position: -200% 0; }\n }\n\n .scaler-events-empty {\n text-align: center;\n padding: 40px 20px;\n color: var(--scaler-text-secondary);\n font-size: 15px;\n }\n\n @media (max-width: 768px) {\n .scaler-events-carousel { padding: 16px; }\n .scaler-event-swiper .swiper-slide { width: calc(100vw - 52px); }\n .scaler-events-nav__btn { display: none; }\n }\n <\/style>\n\n\n\n <div class=\"scaler-events-carousel js-scaler-carousel\">\n\n <template class=\"js-event-card-template\">\n <div class=\"swiper-slide\">\n <article class=\"scaler-event-card\">\n <div class=\"scaler-event-card__header\">\n <div class=\"scaler-event-card__image-wrapper\">\n <img decoding=\"async\" src=\"\" alt=\"\" class=\"scaler-event-card__image\" loading=\"lazy\">\n <div class=\"scaler-event-card__img-footer\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M11.199 7.957c-3.086 0-5.588 2.502-5.588 5.588s2.502 5.588 5.588 5.588c3.086 0 5.588-2.502 5.588-5.588s-2.502-5.588-5.588-5.588zM3.611 13.545c0-4.191 3.397-7.588 7.588-7.588s7.588 3.397 7.588 7.588c0 4.191-3.397 7.588-7.588 7.588s-7.588-3.397-7.588-7.588z\"\/><path d=\"M21.090 6.369h0.003c1.989 0 3.897 0.79 5.303 2.197s2.197 3.314 2.197 5.303-0.79 3.897-2.197 5.303c-1.406 1.407-3.314 2.197-5.303 2.197-0.552 0-1-0.448-1-1s0.448-1 1-1c1.459 0 2.858-0.579 3.889-1.611s1.611-2.43 1.611-3.889c0-1.459-0.58-2.858-1.611-3.889s-2.429-1.611-3.888-1.611c-0.507 0.002-1.013 0.070-1.502 0.203-0.533 0.145-1.082-0.17-1.227-0.703s0.17-1.082 0.703-1.227c0.659-0.179 1.339-0.27 2.021-0.272z\"\/><path d=\"M10.906 21.364c-1.605 0-3.186 0.386-4.61 1.126s-2.649 1.811-3.572 3.124c-0.318 0.452-0.941 0.561-1.393 0.243s-0.561-0.941-0.243-1.393c1.107-1.576 2.577-2.862 4.286-3.749s3.606-1.351 5.532-1.351 3.823 0.463 5.532 1.351c1.709 0.888 3.179 2.174 4.286 3.749 0.318 0.452 0.209 1.076-0.243 1.393s-1.076 0.209-1.393-0.243c-0.923-1.313-2.148-2.385-3.572-3.124s-3.005-1.126-4.61-1.126z\"\/><\/svg>\n <span class=\"js-event-count\"><\/span>\n <\/div>\n <div class=\"scaler-event-card__live-badge\" style=\"display: none;\">Live<\/div>\n <\/div>\n <div class=\"scaler-event-card__text\">\n <h3 class=\"scaler-event-card__title\"><\/h3>\n <div class=\"scaler-event-card__speaker js-event-speaker\"><\/div>\n <\/div>\n <\/div>\n\n <div class=\"scaler-event-card__takeaways js-event-takeaways\"><\/div>\n\n <div class=\"scaler-event-card__spacer\"><\/div>\n\n <div class=\"scaler-event-card__meta\">\n <span class=\"js-event-date\"><\/span>\n <span class=\"scaler-event-card__duration\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M16 6.333c6.443 0 11.667 5.223 11.667 11.667s-5.223 11.667-11.667 11.667c-6.443 0-11.667-5.223-11.667-11.667v0c0.008-6.44 5.226-11.659 11.666-11.667h0.001zM16 28.333c5.707 0 10.333-4.626 10.333-10.333s-4.626-10.333-10.333-10.333c-5.707 0-10.333 4.626-10.333 10.333v0c0.007 5.704 4.629 10.326 10.333 10.333h0.001z\"\/><path d=\"M16 18.667c-0 0-0.001 0-0.001 0-0.184 0-0.35-0.074-0.471-0.195v0c-0.121-0.121-0.195-0.287-0.195-0.471s0.075-0.351 0.195-0.471l4.951-4.952c0.121-0.121 0.287-0.195 0.471-0.195s0.351 0.075 0.471 0.195v0c0.121 0.121 0.195 0.287 0.195 0.471s-0.075 0.351-0.195 0.471l-4.949 4.951c-0.121 0.121-0.288 0.196-0.472 0.196v0z\"\/><path d=\"M19 3.667h-6c-0.368 0-0.667-0.298-0.667-0.667v0c0-0.368 0.298-0.667 0.667-0.667v0h6c0.368 0 0.667 0.298 0.667 0.667v0c0 0.368-0.298 0.667-0.667 0.667v0z\"\/><\/svg>\n <span class=\"js-event-duration\"><\/span>\n <\/span>\n <\/div>\n\n <a href=\"#\" class=\"scaler-event-card__cta js-event-cta\">Register with 1-Click<\/a>\n <\/article>\n <\/div>\n <\/template>\n\n <div class=\"scaler-events-carousel__inner\">\n <header class=\"scaler-events-header\">\n <span class=\"scaler-events-header__icon\">\n <svg fill=\"currentColor\" viewBox=\"0 0 33 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M6.406 6.8c-0.11 0-0.2 0.090-0.2 0.2v20c0 0.11 0.090 0.2 0.2 0.2h20c0.11 0 0.2-0.090 0.2-0.2v-20c0-0.11-0.090-0.2-0.2-0.2h-20zM4.606 7c0-0.994 0.806-1.8 1.8-1.8h20c0.994 0 1.8 0.806 1.8 1.8v20c0 0.994-0.806 1.8-1.8 1.8h-20c-0.994 0-1.8-0.806-1.8-1.8v-20z\"\/><path d=\"M22.406 3.2c0.442 0 0.8 0.358 0.8 0.8v4c0 0.442-0.358 0.8-0.8 0.8s-0.8-0.358-0.8-0.8v-4c0-0.442 0.358-0.8 0.8-0.8z\"\/><path d=\"M10.406 3.2c0.442 0 0.8 0.358 0.8 0.8v4c0 0.442-0.358 0.8-0.8 0.8s-0.8-0.358-0.8-0.8v-4c0-0.442 0.358-0.8 0.8-0.8z\"\/><path d=\"M4.606 12c0-0.442 0.358-0.8 0.8-0.8h22c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-22c-0.442 0-0.8-0.358-0.8-0.8z\"\/><path d=\"M21.488 16.451c0.303 0.322 0.288 0.828-0.034 1.131l-5.838 5.5c-0.309 0.291-0.791 0.29-1.099-0.002l-3.162-3c-0.321-0.304-0.334-0.81-0.030-1.131s0.81-0.334 1.131-0.030l2.614 2.48 5.287-4.981c0.322-0.303 0.828-0.288 1.131 0.034z\"\/><\/svg>\n <\/span>\n <h2 class=\"scaler-events-header__title\"><span class=\"ez-toc-section\" id=\"live-masterclasses\"><\/span>Live Masterclasses<span class=\"ez-toc-section-end\"><\/span><\/h2>\n <\/header>\n\n <div class=\"scaler-events-swiper-wrap\">\n <div class=\"swiper scaler-event-swiper\">\n <div class=\"swiper-wrapper scaler-events-wrapper\"><\/div>\n <div class=\"swiper-pagination scaler-events-pagination\"><\/div>\n <\/div>\n\n <button class=\"scaler-events-nav__btn scaler-nav-prev\" aria-label=\"Previous slide\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(-90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"\/><\/svg>\n <\/button>\n <button class=\"scaler-events-nav__btn scaler-nav-next\" aria-label=\"Next slide\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"\/><\/svg>\n <\/button>\n <\/div>\n\n <a class=\"scaler-events-viewall\" href=\"https:\/\/www.scaler.com\/events\/\">\n <span>View All Events<\/span>\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"\/><\/svg>\n <\/a>\n <\/div>\n <\/div>\n\n <script>\n document.addEventListener(\"DOMContentLoaded\", () => {\n\n const monthNames = [\"January\",\"February\",\"March\",\"April\",\"May\",\"June\",\"July\",\"August\",\"September\",\"October\",\"November\",\"December\"];\n\n function formatTime(st) {\n const t = new Date(st);\n const date = t.getDate();\n const month = monthNames[t.getMonth()];\n const year = t.getFullYear();\n const h = t.getHours();\n const m = String(t.getMinutes()).padStart(2, \"0\");\n let time;\n if (h === 0) time = `12:${m} AM`;\n else if (h === 12) time = `12:${m} PM`;\n else if (h > 12) time = `${h - 12}:${m} PM`;\n else time = `${h}:${m} AM`;\n return `${date} ${month} ${year} | ${time}`;\n }\n\n const carouselInstances = document.querySelectorAll('.js-scaler-carousel');\n\n carouselInstances.forEach(container => {\n\n if(container.dataset.initialized === \"true\") return;\n container.dataset.initialized = \"true\";\n\n const swiperElement = container.querySelector(\".scaler-event-swiper\");\n const swiperWrapper = container.querySelector(\".scaler-events-wrapper\");\n const template = container.querySelector(\".js-event-card-template\");\n const nextBtn = container.querySelector(\".scaler-nav-next\");\n const prevBtn = container.querySelector(\".scaler-nav-prev\");\n const paginationEl = container.querySelector(\".scaler-events-pagination\");\n\n if (!swiperWrapper || !template) {\n console.error(\"Scaler Carousel: Missing required elements inside container\");\n return;\n }\n\n const swiper = new Swiper(swiperElement, {\n slidesPerView: \"auto\",\n spaceBetween: 16,\n grabCursor: true,\n observer: true,\n observeParents: true,\n pagination: {\n el: paginationEl,\n clickable: true,\n dynamicBullets: true\n },\n navigation: {\n nextEl: nextBtn,\n prevEl: prevBtn\n },\n });\n\n function showSkeletons(count = 4) {\n swiperWrapper.innerHTML = \"\";\n for (let i = 0; i < count; i++) {\n const clone = template.content.cloneNode(true);\n const card = clone.querySelector(\".scaler-event-card\");\n card.classList.add(\"scaler-event-card--skeleton\");\n swiperWrapper.appendChild(clone);\n }\n swiper.update();\n }\n\n function renderEvents(events) {\n swiperWrapper.innerHTML = \"\";\n\n if (events.length === 0) {\n swiperWrapper.innerHTML = `<div class=\"scaler-events-empty\">No upcoming masterclasses found.<\/div>`;\n return;\n }\n\n const pathParts = location.pathname.split(\"\/\").filter(Boolean);\n const currentSlug = pathParts.length > 0 ? pathParts[pathParts.length - 1] : \"homepage\";\n\n events.forEach(event => {\n const attr = event.attributes;\n const cd = attr.custom_data || {};\n const clone = template.content.cloneNode(true);\n const joinUrl = `\/event\/${attr.slug}\/?utm_source=blog&utm_medium=master_class&utm_content=${currentSlug}`;\n\n \/\/ Instructor headshot\n const instructor = (cd.instructors && cd.instructors[0]) || {};\n const img = clone.querySelector(\".scaler-event-card__image\");\n img.src = instructor.media_url || attr.image_url ||\n \"https:\/\/images.unsplash.com\/photo-1531891437562-4301cf35b7e4?w=200&h=200&fit=crop\";\n img.alt = instructor.name || attr.instructor_name || attr.title;\n\n \/\/ Title + speaker\n clone.querySelector(\".scaler-event-card__title\").textContent = attr.title;\n clone.querySelector(\".js-event-speaker\").textContent = instructor.name || attr.instructor_name || \"\";\n\n \/\/ Live chip vs attendee count\n const isLive = attr.status === \"current\";\n const footer = clone.querySelector(\".scaler-event-card__img-footer\");\n const liveBadge = clone.querySelector(\".scaler-event-card__live-badge\");\n if (isLive) {\n footer.style.display = \"none\";\n liveBadge.style.display = \"flex\";\n liveBadge.textContent = \"Live\";\n } else if (attr.registration_count) {\n clone.querySelector(\".js-event-count\").textContent = attr.registration_count;\n } else {\n footer.style.display = \"none\";\n }\n\n \/\/ Takeaways (up to 2, Know More on the last)\n const takeaways = (cd.takeaways || []).slice(0, 2);\n const listEl = clone.querySelector(\".js-event-takeaways\");\n listEl.innerHTML = \"\";\n takeaways.forEach((t, idx) => {\n const item = document.createElement(\"div\");\n item.className = \"scaler-event-card__takeaway-item\";\n const dot = document.createElement(\"span\");\n dot.className = \"scaler-event-card__dot\";\n const txt = document.createElement(\"span\");\n txt.className = \"scaler-event-card__takeaway-text\";\n txt.textContent = t;\n item.appendChild(dot);\n item.appendChild(txt);\n if (idx === takeaways.length - 1) {\n const km = document.createElement(\"a\");\n km.className = \"scaler-event-card__knowmore\";\n km.textContent = \"Know More\";\n km.href = joinUrl;\n item.appendChild(km);\n }\n listEl.appendChild(item);\n });\n\n \/\/ Date + duration\n clone.querySelector(\".js-event-date\").textContent = formatTime(attr.start_time);\n let hrs = \"\";\n if (attr.duration) {\n hrs = (attr.duration \/ 60).toFixed(2);\n } else if (attr.end_time) {\n hrs = (Math.abs(new Date(attr.start_time) - new Date(attr.end_time)) \/ 3600000).toFixed(2);\n }\n clone.querySelector(\".js-event-duration\").textContent = hrs ? `${hrs} Hrs` : \"\";\n\n \/\/ CTA\n clone.querySelector(\".js-event-cta\").href = joinUrl;\n\n swiperWrapper.appendChild(clone);\n });\n\n swiper.update();\n swiper.slideTo(0);\n }\n\n async function fetchEvents() {\n try {\n showSkeletons();\n const res = await fetch(\n \"https:\/\/www.scaler.com\/api\/v4\/events?event_type[]=company&distributor=scaler&type=upcoming&serializer_mode=L2&limit=8&program[]=software_development&program[]=data_science&program[]=devops&program[]=ai_ml\"\n );\n const json = await res.json();\n const events = json.data || [];\n renderEvents(events);\n } catch (error) {\n console.error(\"Failed to load events:\", error);\n if(swiperWrapper) swiperWrapper.innerHTML = `<div class=\"scaler-events-empty\">Failed to load events.<\/div>`;\n }\n }\n\n fetchEvents();\n });\n });\n <\/script>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>3. Named Entity Recognition (NER) Pipeline with spaCy<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Named Entity Recognition helps extract specific details from unstructured text. In this project, use spaCy to find entities such as people, organisations, locations, dates, and monetary values in documents. Start with spaCy’s pre-trained NER pipeline and run it on sample documents to see which entities it identifies correctly and where it makes mistakes.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For a more advanced version, train the pipeline on your own labelled examples. This works particularly well with domain-specific documents, where the standard model may not recognise the terms and entities you need. Recruitment resumes, financial reports, and legal documents are good examples.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, spaCy, a collection of sample documents, and labelled examples if you plan to train a custom NER model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/spacy.io\/api\/entityrecognizer\" target=\"_blank\" rel=\"noopener\">Named Entity Recognition<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Named Entity Recognition, spaCy, entity extraction, text preprocessing, custom NER training.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>4. Keyword Extraction Tool<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Reading a long article or report can make it hard to identify what the text is mainly about. Keyword extraction can pull out the important terms and phrases, which can then be used to organise documents, improve search, or generate a quick overview of the content.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For this project, start with TF-IDF to rank terms based on how important they are in one document compared with the rest of the collection. Then try YAKE! This is an unsupervised keyword-extraction method that looks at features such as word frequency, position, and context within a single document. Run both methods on news articles, blog posts, or research papers and compare the keywords they produce.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, a collection of documents, scikit-learn for TF-IDF, and the YAKE! Python library.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/github.com\/INESCTEC\/yake\" target=\"_blank\" rel=\"noopener\">YAKE! – GitHub<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Keyword extraction, TF-IDF, YAKE!, text ranking, document analysis.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>5. News Search Engine<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">When you search for a news topic, you expect the most relevant articles to appear first. A basic search engine has to take the words in your query, compare them with a collection of articles, and decide which documents deserve the highest rank.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For this nlp project, use the BBC News Classification Dataset and preprocess the articles with steps such as tokenization, stop-word removal, and stemming. Use BM25 with the rank_bm25 library to score the articles against a user’s query and return the highest-ranking results. BM25 is related to TF-IDF, but it also accounts for factors such as term frequency and document length when calculating relevance.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Try different search queries and check whether the highest-ranked articles are actually relevant. You can also experiment with the preprocessing steps and see how they affect the results.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, pandas, NLTK, Gensim, rank_bm25, and the BBC News Classification Dataset.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.scaler.com\/topics\/nlp\/build-news-search-engine-using-nlp\/\">Build a News Search Engine with NLP<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Information retrieval, text preprocessing, BM25, document ranking, query matching, search pipelines.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>6. Resume Parser<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Resumes come in different formats and use different headings for the same information. One may have a \u201cTechnical Skills\u201d section, another may use \u201cCore Competencies,\u201d while education and work experience can be arranged in completely different ways. A resume parser needs to find these details and turn them into consistent fields such as name, skills, education, job titles, and experience.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Here, extract text from resumes and use spaCy’s Named Entity Recognition (NER) to identify the information you need. Start with a pre-trained NER model and train it further with labelled resume data for fields that aren’t recognised correctly. You can then store the extracted details in a structured format for candidate search or skill matching.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, spaCy, resume documents, and labelled resume data for custom entities.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Explore the source code:<\/strong> <a href=\"https:\/\/github.com\/topics\/resume-parser\" target=\"_blank\" rel=\"noopener\">GitHub Resume Parser Projects<\/a> <\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Resume parsing, Named Entity Recognition, information extraction, custom NER training, structured data extraction.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"era-2-embeddings-transformers-projects-7-12\"><\/span><strong>Era 2: Embeddings & Transformers (Projects 7-12)<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Imagine searching for <strong>“automobile”<\/strong> and missing a relevant document simply because it used the word <strong>“car”<\/strong> instead. Classical NLP often relied on exact word matches or handcrafted features, which sometimes made it difficult to capture the meaning behind text when another word would be used instead.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Embeddings and transformer models made it easier for NLP systems to work with meaning and context. Embeddings help a model recognise that words used in similar situations can be related, even when the words themselves are different. Transformers go a step further by looking at the words around a term, which helps the model understand what that term means in a particular sentence. These techniques are now used for tasks such as sentiment analysis, summarization, question answering, and semantic search.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The nlp project ideas in this section cover word embeddings, BERT fine-tuning, text summarization, question answering, emotion detection, and toxic-comment classification. You’ll use pretrained models from the Hugging Face ecosystem for the transformer-based projects. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">If you’re new to deep learning, <a href=\"https:\/\/www.scaler.com\/topics\/course\/keras-tensorflow-for-deep-learning\/\"><strong>Scaler’s free Keras & TensorFlow<\/strong><\/a> course covers neural networks, transfer learning, and transformer architectures before you start.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>7. Word Embedding Explorer<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">As we mentioned previously, consider these words, \u201ccar\u201d and \u201cautomobile.\u201d A traditional text classifier may treat them as two separate words, but word embeddings can represent them as vectors that reflect how they are used in language. Words that appear in similar contexts tend to have similar positions in this vector space.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For this, use a pretrained Word2Vec, GloVe, or FastText model and create an explorer where you can enter a word, find its closest terms, and see how related words are grouped together. Try word-analogy tasks such as \u201cking\u201d \u2212 \u201cman\u201d + \u201cwoman\u201d and visualise the embeddings in two or three dimensions using PCA or t-SNE.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, a pretrained Word2Vec, GloVe, or FastText model, and a library such as Gensim for working with word embeddings.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.scaler.com\/topics\/tensorflow\/tensorflow-word-embeddings\/\">Word Embeddings with TensorFlow<\/a><\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>8. Fine-Tune BERT for Sentiment Analysis<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A pretrained BERT model already contains language patterns learned from large amounts of text. In this project, adapt BERT to sentiment analysis using the <a href=\"https:\/\/www.tensorflow.org\/datasets\/catalog\/imdb_reviews\" target=\"_blank\" rel=\"noopener\">IMDB movie-review dataset<\/a>, where reviews are labelled as positive or negative. Fine-tuning allows the model to adjust its existing language representations to recognise these sentiment labels.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Use the <a href=\"https:\/\/www.scaler.com\/topics\/tensorflow\/\">TensorFlow tutorial<\/a> to load the dataset, preprocess the text for BERT, add a classification layer, and fine-tune the model on the labelled reviews. Once trained, test it on unseen reviews and compare its predictions with the TF-IDF-based sentiment classifier from Project 2 using accuracy, precision, recall, and F1-score.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, TensorFlow, TensorFlow Hub, TensorFlow Text, the IMDB dataset, and a pretrained BERT model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.tensorflow.org\/text\/tutorials\/classify_text_with_bert\" target=\"_blank\" rel=\"noopener\">Classify Text with BERT – TensorFlow<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> BERT fine-tuning, transfer learning, text preprocessing, sentiment classification, model evaluation.<\/p>\n\n\n<meta charset=\"UTF-8\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n\n <!-- Source Sans Pro (topics font) -->\n <link href=\"https:\/\/fonts.googleapis.com\/css2?family=Source+Sans+3:wght@400;600;700&display=swap\" rel=\"stylesheet\">\n <link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.css\">\n <script src=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.js\"><\/script>\n\n <style>\n :root {\n --cl-primary: #005bb3;\n --cl-brand-blue: #255aad;\n --cl-card-bg: #ffffff;\n --cl-surface: #f6f7f8;\n --cl-border: #d6dee5;\n --cl-text-primary: #1a2c47; \/* neutral-scale-12 *\/\n --cl-text-secondary: #61738e; \/* neutral-scale-9 *\/\n --cl-radius: 16px;\n --cl-radius-sm: 8px;\n --cl-shadow: 0 8px 16px 0 rgba(50, 71, 102, 0.16);\n }\n\n .scaler-courses * { box-sizing: border-box; }\n\n .scaler-courses {\n font-family: \"Source Sans 3\", \"Source Sans Pro\", \"Roboto\", Arial, sans-serif;\n padding: 24px;\n width: 100%;\n }\n .scaler-courses__inner {\n max-width: 1000px;\n margin: 0 auto;\n position: relative;\n width: 100%;\n }\n\n \/* Header *\/\n .scaler-courses__header {\n display: flex;\n align-items: center;\n gap: 8px;\n margin-bottom: 8px;\n padding: 8px 0;\n }\n .scaler-courses__header-icon { display: inline-flex; color: var(--cl-text-primary); }\n .scaler-courses__header-icon svg { width: 24px; height: 24px; }\n .scaler-courses__title {\n font-size: 18px;\n font-weight: 700;\n color: var(--cl-text-primary);\n line-height: 1.3;\n margin: 0;\n }\n\n \/* Swiper *\/\n .scaler-courses__swiper-wrap { position: relative; }\n .scaler-courses .swiper { width: 100%; padding: 4px; }\n .scaler-courses-swiper .swiper-slide {\n width: 296px;\n height: auto;\n display: flex;\n }\n\n \/* Card *\/\n .scaler-course-card {\n display: flex;\n flex-direction: column;\n width: 100%;\n height: 100%;\n min-height: 336px;\n background: var(--cl-card-bg);\n border: 1px solid var(--cl-border);\n border-radius: var(--cl-radius);\n overflow: hidden;\n text-decoration: none;\n cursor: pointer;\n transition: box-shadow 0.2s ease;\n }\n .scaler-course-card:hover { box-shadow: 0 2px 12px rgba(50, 71, 102, 0.12); }\n\n .scaler-course-card__image {\n width: 100%;\n height: 128px;\n min-height: 128px;\n object-fit: cover;\n background: var(--cl-surface);\n display: block;\n }\n\n .scaler-course-card__detail {\n display: flex;\n flex-direction: column;\n flex: 1 1 auto;\n padding: 16px 16px 8px;\n }\n .scaler-course-card__heading,\n .scaler-course-card__heading:link,\n .scaler-course-card__heading:visited {\n font-size: 20px;\n font-weight: 600;\n line-height: 1.4;\n color: var(--cl-text-primary) !important;\n margin: 0 0 8px;\n white-space: nowrap;\n overflow: hidden;\n text-overflow: ellipsis;\n }\n .scaler-course-card__instructor {\n font-size: 14px;\n line-height: 1.86;\n color: var(--cl-text-primary);\n white-space: nowrap;\n overflow: hidden;\n text-overflow: ellipsis;\n }\n .scaler-course-card__instructor b { font-weight: 700; }\n .scaler-course-card__desc {\n flex-grow: 1;\n margin-top: 8px;\n margin-bottom: 12px;\n font-size: 16px;\n line-height: 1.5;\n color: var(--cl-text-secondary);\n display: -webkit-box;\n -webkit-line-clamp: 2;\n -webkit-box-orient: vertical;\n overflow: hidden;\n }\n\n .scaler-course-card__footer {\n display: flex;\n align-items: center;\n justify-content: space-between;\n width: 100%;\n }\n .scaler-course-card__stat {\n display: flex;\n align-items: center;\n gap: 4px;\n font-size: 14px;\n color: var(--cl-text-primary);\n }\n .scaler-course-card__stat svg { width: 18px; height: 18px; flex-shrink: 0; }\n .scaler-course-card__stat img { width: 20px; height: 20px; flex-shrink: 0; }\n\n \/* Side arrows *\/\n .scaler-courses__nav {\n position: absolute !important;\n top: 50% !important;\n transform: translateY(-50%) !important;\n z-index: 2;\n width: 46px !important; height: 46px !important; min-width: 46px;\n display: flex !important; align-items: center; justify-content: center;\n padding: 0 !important;\n background: var(--cl-card-bg) !important;\n border: none !important;\n border-radius: var(--cl-radius-sm) !important;\n box-shadow: var(--cl-shadow) !important;\n color: var(--cl-text-secondary) !important;\n cursor: pointer;\n transition: opacity 0.2s ease, visibility 0.2s ease;\n }\n .scaler-courses__nav--prev { left: 8px !important; }\n .scaler-courses__nav--next { right: 8px !important; }\n .scaler-courses__nav svg { width: 20px !important; height: 20px !important; fill: var(--cl-text-secondary) !important; }\n .scaler-courses__nav.swiper-button-disabled { opacity: 0; visibility: hidden; pointer-events: none; }\n\n .scaler-courses .swiper-button-prev,\n .scaler-courses .swiper-button-next,\n .scaler-courses .swiper-pagination { display: none; }\n\n \/* Skeleton *\/\n .scaler-course-card--skeleton { pointer-events: none; }\n .scaler-course-card--skeleton .scaler-course-card__image,\n .scaler-course-card--skeleton .scaler-course-card__heading,\n .scaler-course-card--skeleton .scaler-course-card__instructor,\n .scaler-course-card--skeleton .scaler-course-card__desc,\n .scaler-course-card--skeleton .scaler-course-card__footer {\n background: linear-gradient(90deg, var(--cl-border) 25%, var(--cl-surface) 50%, var(--cl-border) 75%);\n background-size: 200% 100%;\n animation: cl-shimmer 1.5s infinite;\n color: transparent !important;\n border-radius: 4px;\n }\n .scaler-course-card--skeleton .scaler-course-card__heading { height: 24px; margin: 0 0 8px; }\n .scaler-course-card--skeleton .scaler-course-card__instructor { height: 16px; }\n .scaler-course-card--skeleton .scaler-course-card__footer { height: 20px; }\n @keyframes cl-shimmer { 0% { background-position: 200% 0; } 100% { background-position: -200% 0; } }\n\n .scaler-courses__empty { display: none; }\n\n @media (max-width: 768px) {\n .scaler-courses { padding: 16px; }\n .scaler-courses-swiper .swiper-slide { width: 260px; }\n .scaler-courses__nav { display: none !important; }\n }\n <\/style>\n\n\n <div class=\"scaler-courses js-scaler-courses\">\n\n <template class=\"js-course-card-template\">\n <div class=\"swiper-slide\">\n <a class=\"scaler-course-card js-course-link\" href=\"#\" target=\"_blank\" rel=\"noopener\">\n <img decoding=\"async\" class=\"scaler-course-card__image js-course-image\" src=\"\" alt=\"\" loading=\"lazy\">\n <div class=\"scaler-course-card__detail\">\n <div class=\"scaler-course-card__heading js-course-title\"><\/div>\n <div class=\"scaler-course-card__instructor js-course-instructor\"><\/div>\n <div class=\"scaler-course-card__desc js-course-desc\"><\/div>\n <div class=\"scaler-course-card__footer\">\n <span class=\"scaler-course-card__stat js-course-star\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M14.83 2.868c0.342-0.239 0.75-0.368 1.168-0.368s0.826 0.128 1.168 0.368c0.338 0.236 0.596 0.569 0.741 0.955l2.759 6.936c0.003 0.007 0.007 0.013 0.013 0.017s0.013 0.007 0.020 0.007l7.402 0.476c0.411 0.036 0.802 0.194 1.124 0.452s0.559 0.607 0.683 1 0.127 0.815 0.011 1.211c-0.116 0.396-0.347 0.748-0.664 1.013l-5.655 4.714c-0.013 0.011-0.022 0.025-0.028 0.040s-0.006 0.032-0.002 0.047l1.821 7.163c0.107 0.395 0.094 0.812-0.037 1.2-0.133 0.391-0.38 0.734-0.71 0.983s-0.727 0.393-1.14 0.413c-0.413 0.020-0.822-0.084-1.175-0.3l-0.015-0.009-6.307-4.005c-0.004-0.003-0.009-0.004-0.014-0.004s-0.009 0.001-0.013 0.004l-5.876 3.721c-0.389 0.235-0.839 0.347-1.292 0.323s-0.889-0.184-1.251-0.459c-0.361-0.275-0.632-0.652-0.777-1.082-0.143-0.426-0.157-0.885-0.038-1.319l1.684-6.629c0.004-0.016 0.003-0.033-0.002-0.049s-0.015-0.030-0.027-0.040l-5.653-4.714c-0.317-0.264-0.548-0.617-0.664-1.013s-0.112-0.817 0.011-1.211c0.124-0.394 0.361-0.742 0.683-1s0.713-0.416 1.124-0.452l0.023-0.002 7.38-0.476c0.007-0 0.014-0.003 0.020-0.007s0.010-0.010 0.013-0.017l0.008-0.020 2.746-6.913c0.145-0.386 0.403-0.719 0.741-0.956z\"\/><\/svg>\n <span class=\"js-course-star-val\"><\/span>\n <\/span>\n <span class=\"scaler-course-card__stat\">\n <img decoding=\"async\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/user-check-dark.svg\" alt=\"Enrolled\" loading=\"lazy\">\n <span class=\"js-course-reg\"><\/span>\n <\/span>\n <\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/template>\n\n <div class=\"scaler-courses__inner\">\n <div class=\"scaler-courses__header\">\n <span class=\"scaler-courses__header-icon\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M26.259 23.289h-20.518c-1.193 0-2.16-0.967-2.16-2.16v0-14.588c0-1.193 0.967-2.16 2.16-2.16v0h20.518c1.193 0 2.16 0.967 2.16 2.16v0 14.588c0 1.193-0.967 2.16-2.16 2.16v0zM5.741 5.501c-0.574 0-1.040 0.466-1.040 1.040v0 14.588c0 0.574 0.466 1.040 1.040 1.040v0h20.518c0.574 0 1.040-0.466 1.040-1.040v0-14.588c0-0.574-0.466-1.040-1.040-1.040v0z\"\/><path d=\"M19.764 27.619h-7.53c-0.309 0-0.56-0.251-0.56-0.56v0c0-0.309 0.251-0.56 0.56-0.56h7.53c0.309 0 0.56 0.251 0.56 0.56v0c0 0.309-0.251 0.56-0.56 0.56v0z\"\/><path d=\"M14.023 9.322c0.116 0 0.223 0.035 0.312 0.096l-0.002-0.001 5.93 3.953c0.151 0.102 0.249 0.272 0.249 0.466s-0.098 0.364-0.247 0.465l-0.002 0.001-5.93 3.953c-0.087 0.059-0.195 0.094-0.31 0.094-0.309 0-0.56-0.25-0.56-0.559v-7.906c0-0.309 0.251-0.56 0.56-0.56v0zM18.943 13.835l-4.36-2.906v5.813z\"\/><\/svg>\n <\/span>\n <h2 class=\"scaler-courses__title\"><span class=\"ez-toc-section\" id=\"free-courses-by-top-scaler-instructors\"><\/span>Free Courses by top Scaler instructors<span class=\"ez-toc-section-end\"><\/span><\/h2>\n <\/div>\n\n <div class=\"scaler-courses__swiper-wrap\">\n <div class=\"swiper scaler-courses-swiper\">\n <div class=\"swiper-wrapper js-courses-wrapper\"><\/div>\n <div class=\"swiper-pagination\"><\/div>\n <\/div>\n\n <button class=\"scaler-courses__nav scaler-courses__nav--prev js-courses-prev\" aria-label=\"Previous\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(-90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"\/><\/svg>\n <\/button>\n <button class=\"scaler-courses__nav scaler-courses__nav--next js-courses-next\" aria-label=\"Next\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"\/><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <script>\n document.addEventListener(\"DOMContentLoaded\", () => {\n const EXTRA_COUNT_ADDED = 1000;\n\n \/\/ Mirrors getInstructorStar(): first token of the headline.\n function getInstructorStar(headline) {\n return (headline || \"\").toString().split(\" \")[0] || \"\";\n }\n\n document.querySelectorAll(\".js-scaler-courses\").forEach(container => {\n if (container.dataset.initialized === \"true\") return;\n container.dataset.initialized = \"true\";\n\n const swiperElement = container.querySelector(\".scaler-courses-swiper\");\n const wrapper = container.querySelector(\".js-courses-wrapper\");\n const template = container.querySelector(\".js-course-card-template\");\n const nextBtn = container.querySelector(\".js-courses-next\");\n const prevBtn = container.querySelector(\".js-courses-prev\");\n\n if (!wrapper || !template) return;\n\n const swiper = new Swiper(swiperElement, {\n slidesPerView: \"auto\",\n spaceBetween: 16,\n grabCursor: true,\n observer: true,\n observeParents: true,\n navigation: { nextEl: nextBtn, prevEl: prevBtn },\n });\n\n const pathParts = location.pathname.split(\"\/\").filter(Boolean);\n const currentSlug = pathParts.length > 0 ? pathParts[pathParts.length - 1] : \"homepage\";\n\n function showSkeletons(count = 4) {\n wrapper.innerHTML = \"\";\n for (let i = 0; i < count; i++) {\n const clone = template.content.cloneNode(true);\n clone.querySelector(\".scaler-course-card\").classList.add(\"scaler-course-card--skeleton\");\n wrapper.appendChild(clone);\n }\n swiper.update();\n }\n\n function renderCourses(courses) {\n wrapper.innerHTML = \"\";\n if (!courses.length) { container.style.display = \"none\"; return; }\n\n courses.forEach(item => {\n const a = item.attributes || {};\n const clone = template.content.cloneNode(true);\n const instructor = (a.instructors_data && a.instructors_data[0]) || {};\n\n clone.querySelector(\".js-course-link\").href =\n `https:\/\/www.scaler.com\/topics\/course\/${a.slug}\/?utm_source=blog&utm_medium=free_courses&utm_content=${currentSlug}`;\n\n const img = clone.querySelector(\".js-course-image\");\n img.src = a.cover_image_thumbnail || a.featured_image || \"\";\n img.alt = a.title || \"\";\n\n clone.querySelector(\".js-course-title\").textContent = a.title || \"\";\n\n const instrEl = clone.querySelector(\".js-course-instructor\");\n if (instructor.name) {\n instrEl.innerHTML = \"A Course by <b><\/b>\";\n instrEl.querySelector(\"b\").textContent = instructor.name;\n } else {\n instrEl.style.display = \"none\";\n }\n\n clone.querySelector(\".js-course-desc\").textContent = a.description || \"\";\n\n \/\/ Star\n const star = getInstructorStar(instructor.star_headline);\n const starWrap = clone.querySelector(\".js-course-star\");\n if (star && star !== \"0\") clone.querySelector(\".js-course-star-val\").textContent = star;\n else starWrap.style.display = \"none\";\n\n \/\/ Enrolled count (+1000 like topics)\n clone.querySelector(\".js-course-reg\").textContent =\n ((a.user_registration_count || 0) + EXTRA_COUNT_ADDED).toLocaleString(\"en-US\");\n\n wrapper.appendChild(clone);\n });\n\n swiper.update();\n swiper.slideTo(0);\n }\n\n async function fetchCourses() {\n try {\n showSkeletons();\n \/\/ Public same-origin endpoint (topics course search). Returns 6 free courses.\n const res = await fetch(\"\/topics\/api\/v1\/search\/courses\", {\n method: \"POST\",\n headers: {\n \"Content-Type\": \"application\/json\",\n \"Accept\": \"application\/json\",\n \"X-Requested-With\": \"XMLHttpRequest\",\n },\n body: JSON.stringify({\n filters: {\n q: \"\",\n page: { limit: 6, current_page: 1 },\n sort: { key: \"user_registration_count\", order: \"desc\" },\n filter_type: null,\n tags: [],\n program: [],\n level: [],\n category_type: [],\n },\n }),\n });\n const json = await res.json();\n const courses = (json && json.data) || [];\n renderCourses(courses);\n } catch (error) {\n console.error(\"Failed to load courses:\", error);\n container.style.display = \"none\";\n }\n }\n\n fetchCourses();\n });\n });\n <\/script>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>9. AI-Powered Text Summarizer<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Newsrooms, legal teams, and analysts often have to work through large amounts of text before they can get to the information they need. A summarizer can shorten that material while retaining the main points.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">One way to build this is with T5, a sequence-to-sequence transformer that generates a summary from the source text. The Hugging Face tutorial uses the BillSum dataset, which pairs legislative documents with human-written summaries. Train T5 on these text-summary pairs, then use the model to generate summaries for documents it has not seen before. ROUGE scores can help compare the generated summaries with the reference summaries.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Hugging Face Transformers, Datasets and Evaluate, the BillSum dataset, and a pretrained T5 model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/huggingface.co\/docs\/transformers\/tasks\/summarization\" target=\"_blank\" rel=\"noopener\">Summarization with Transformers – Hugging Face<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Text summarization, sequence-to-sequence models, T5, transformer fine-tuning, text generation, ROUGE evaluation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>10. Context-Aware Question Answering System<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A question-answering system works with a passage and a question, then identifies the part of the passage that contains the answer. For example, given a product manual and the question \u201cHow long does the battery last?\u201d, the model should locate the relevant words in the manual instead of generating an answer from scratch.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Use DistilBERT and the SQuAD dataset to train an extractive question-answering model. The dataset provides a question, its context, and the exact answer span within that context. During preprocessing, you’ll also need to handle longer passages that exceed the model’s input limit. After fine-tuning, test the model with new questions and check whether it identifies the correct answer span.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Hugging Face Transformers, Datasets and Evaluate, the SQuAD dataset, and a pretrained DistilBERT model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/huggingface.co\/docs\/transformers\/en\/tasks\/question_answering\" target=\"_blank\" rel=\"noopener\">Question Answering with Transformers – Hugging Face<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Extractive question answering, DistilBERT fine-tuning, tokenization, context handling, answer extraction, model evaluation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>11. Emotion Detection from Text<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">\u201cPositive\u201d and \u201cnegative\u201d as answers are not enough sometimes. A customer review can show frustration, relief, excitement, or disappointment even when the overall sentiment appears similar. An emotion detection model looks at the text more closely and assigns it to specific emotion categories.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For this project, train a transformer model to recognise multiple emotions from labelled text. GoEmotions is one dataset you can use for this, with 27 emotion categories and a neutral label. Some examples carry more than one emotion, so the project introduces multi-label classification alongside transformer-based text classification.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Fine-tune a model such as BERT or RoBERTa, then test it on new text and check precision, recall, and F1-score for the different emotion categories.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Hugging Face Transformers, a labelled emotion dataset such as GoEmotions, and a pretrained transformer model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/data:<\/strong> <a href=\"https:\/\/huggingface.co\/papers\/2005.00547\" target=\"_blank\" rel=\"noopener\">GoEmotions Dataset – Hugging Face<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Emotion classification, multi-label classification, transformer fine-tuning, Hugging Face Transformers, model evaluation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>12. Toxic Comment Detection<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Online platforms receive far more comments than people can review manually. A moderation system can flag comments that contain toxic, abusive, or threatening language so they can be reviewed or handled automatically. The difficult part is that the model has to deal with different types of harmful content, and getting a prediction wrong can affect the person who wrote the comment or the person reading it.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You can use the Jigsaw Toxic Comment dataset, which contains Wikipedia comments labelled for categories such as toxic, severe toxic, obscene, threat, insult, and identity hate. Train a transformer-based text classifier on these labels and test it on comments the model has not seen before. Since the categories are not mutually exclusive, this is a multi-label classification problem. Evaluate each label with precision, recall, and F1-score instead of relying on overall accuracy.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Hugging Face Transformers, the Jigsaw Toxic Comment dataset, and a pretrained transformer model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/data:<\/strong> <a href=\"https:\/\/huggingface.co\/datasets\/google\/jigsaw_toxicity_pred\" target=\"_blank\" rel=\"noopener\">Jigsaw Toxic Comment Dataset – Hugging Face<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Toxicity detection, multi-label classification, transformer fine-tuning, content moderation, model evaluation.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"era-3-llm-applications-projects-13-18\"><\/span><strong>Era 3: LLM Applications (Projects 13-18)<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">What happens when an LLM needs information that isn’t part of its training data? It can still produce an answer, but that answer may not be based on the information you actually wanted it to use. This is where techniques such as semantic search and Retrieval-Augmented Generation (RAG) become useful. They allow an application to find relevant information and provide it to the model before it generates a response.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">LLM applications also need to handle tasks that go beyond answering questions. They may have to summarise information from several sources, extract specific details into a fixed format, measure the quality of their responses, or prevent certain outputs. The llm nlp projects in this section cover semantic search, RAG, multi-document summarization, structured extraction, prompt evaluation, and guardrails, giving you a look at the different pieces involved in building LLM-based applications.<\/p>\n\n\n\n<!DOCTYPE html>\n<html>\n <head>\n <title>Hello World!<\/title>\n <link rel=\"preconnect\" href=\"https:\/\/fonts.googleapis.com\">\n <link rel=\"preconnect\" href=\"https:\/\/fonts.gstatic.com\" crossorigin>\n <link href=\"https:\/\/fonts.googleapis.com\/css2?family=Lato:wght@400;600;700&display=swap\" rel=\"stylesheet\">\n <style>\n .iitr_banner_container {\n font-family: lato;\n display: flex;\n flex-direction: row;\n justify-content: space-between;\n border-radius: 16px;\n background: linear-gradient(88deg, #19000F 24.45%, #66003F 83.33%);\n position: relative;\n\n @media (max-width: 768px) {\n min-height: 450px;\n overflow: hidden;\n flex-direction: column;\n }\n }\n .iitr_banner_content {\n display: flex;\n flex-direction: column;\n align-items: flex-start;\n justify-content: center;\n padding: 20px;\n max-width: 50%;\n\n @media (max-width: 768px) {\n max-width: 100%;\n }\n }\n .iitr_banner_title {\n font-size: 24px;\n font-weight: bold;\n color: #FFFFFF;\n\n @media (max-width: 768px) {\n font-size: 20px;\n }\n }\n .iitr_banner_title_highlight {\n color: #FF0071;\n }\n .iitr_banner_subtitle {\n font-size: 14px;\n color: #FFFFFF;\n margin: 10px 0;\n }\n .iitr_banner_btn {\n display: flex;\n justify-content: center;\n align-items: center;\n padding: 8px 48px;\n background-color: #F8F9F9;\n border-radius: 8px;\n border: 1px solid #E3E8E8;\n font-size: 1.4rem;\n font-weight: 600;\n color: #0D3231;\n text-decoration: none;\n margin-top: 16px;\n\n @media (max-width: 768px) {\n padding: 8px 32px;\n }\n }\n .iitr_banner_image {\n position: absolute;\n bottom: 0;\n right: 0;\n\n @media (max-width: 768px) {\n right: auto;\n object-fit: cover;\n min-width: 100%\n }\n }\n .iitr_banner_image_logo {\n margin-bottom: 16px;\n \n @media (max-width: 768px) {\n width: 240px;\n }\n }\n\n \/* Responsive visibility utilities *\/\n .show-in-mobile {\n display: none;\n }\n .hide-in-mobile {\n display: block;\n }\n\n \/* Mobile breakpoint (768px and below) *\/\n @media (max-width: 768px) {\n .show-in-mobile {\n display: block;\n }\n .hide-in-mobile {\n display: none;\n }\n }\n <\/style>\n <\/head>\n <body>\n <div class=\"iitr_banner_container\">\n <div class=\"iitr_banner_content\">\n <img decoding=\"async\" src=\"https:\/\/d2beiqkhq929f0.cloudfront.net\/public_assets\/assets\/000\/176\/281\/original\/Frame_1430102419.svg?1769058073\" class=\"iitr_banner_image_logo\" \/>\n <div class=\"iitr_banner_title\">\n AI Engineering Course Advanced Certification by \n <span class=\"iitr_banner_title_highlight\">\n IIT-Roorkee CEC\n <\/span>\n <\/div>\n <div class=\"iitr_banner_subtitle\">\n A hands on AI engineering program covering Machine Learning, Generative AI, and LLMs – designed for working professionals & delivered by IIT Roorkee in collaboration with Scaler.\n <\/div>\n <a class=\"iitr_banner_btn\" href=\"#\" id=\"iitr_banner_btn\">Enrol Now<\/a>\n <\/div>\n <!-- Desktop Image -->\n <img decoding=\"async\" class=\"iitr_banner_image hide-in-mobile\" src=\"https:\/\/d2beiqkhq929f0.cloudfront.net\/public_assets\/assets\/000\/176\/282\/original\/iitr_2.svg?1769058132\" \/>\n <!-- Mobile Image -->\n <img decoding=\"async\" class=\"iitr_banner_image show-in-mobile\" src=\"https:\/\/d2beiqkhq929f0.cloudfront.net\/public_assets\/assets\/000\/176\/283\/original\/iitr_2_%281%29.svg?1769059469\" \/>\n <\/div>\n <script>\n document.addEventListener(\"DOMContentLoaded\", () => {\n const pathParts = location.pathname.split(\"\/\").filter(Boolean);\n const currentSlug = pathParts.length > 0 ? pathParts[pathParts.length - 1] : \"homepage\";\n const url = `https:\/\/www.scaler.com\/iit-roorkee-advanced-ai-engineering-course?utm_source=blog&utm_medium=iit_roorkee&utm_content=${currentSlug}`;\n const btns = document.querySelectorAll(\".iitr_banner_btn\");\n btns.forEach(btn => {\n btn.href = url;\n });\n });\n <\/script>\n <\/body>\n<\/html>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>13. Semantic Search over Documents<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A search for <strong>\u201cvacation rules\u201d<\/strong> may miss a document titled <strong>\u201cannual leave policy\u201d<\/strong> if the system only looks for matching words. Semantic search can connect the two by looking at the meaning of the query and the text being searched.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For this nlp project, create a collection of documents in Elasticsearch and use its semantic_text field to generate and store semantic representations of the content. Convert the user’s query into the same semantic space and retrieve documents based on their similarity to the query. You can also combine semantic search with traditional keyword matching to see how hybrid search affects the results.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Try queries that use different wording from the documents and check whether the relevant results still appear near the top.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python or Elasticsearch’s query tools, an Elasticsearch instance, a collection of documents, and a model for generating embeddings.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.elastic.co\/docs\/solutions\/search\/get-started\/semantic-search\" target=\"_blank\" rel=\"noopener\">Get Started with Semantic Search – Elastic<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Text embeddings, semantic search, vector search, Elasticsearch, similarity retrieval, hybrid search.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>14. Build a RAG Chatbot Using Your Own Documents<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">What if you want a chatbot to answer questions using information from your own documents? A regular LLM may not have access to that information, even if it can produce a convincing response. A RAG system handles this by finding relevant content first and giving it to the model as context.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You can start by learning how AI chatbots work with this guide to <a href=\"https:\/\/www.scaler.com\/topics\/artificial-intelligence-tutorial\/chatbots-in-ai\/\">chatbots in AI<\/a>, then build a RAG chatbot that works with your own documents. The documents are indexed so the system can search them when a question comes in. The relevant sections are then passed to the LLM, which uses them to generate the response. You can also include citations so users can see where the answer came from.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Microsoft’s sample follows this structure with Azure AI Search for indexing and retrieval and Azure OpenAI for generating responses. It includes sample data and a working Python application, so you can follow the same workflow before replacing the sample documents with your own.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Azure OpenAI, Azure AI Search, a collection of documents, and an embedding model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/learn.microsoft.com\/en-us\/samples\/azure-samples\/azure-search-openai-demo\/azure-search-openai-demo\/\" target=\"_blank\" rel=\"noopener\">RAG Chat App with Your Data – Microsoft Learn<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> RAG architecture, document chunking, embeddings, retrieval, vector search, prompt orchestration, grounded responses.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>15. Multi-Document Summarizer<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Imagine you have several articles covering the same event. Each article may contain information that the others don’t, while some points may be repeated across all of them. Reading every article gives you the details, but you may still want one summary instead of going through each article separately.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Build a summarizer that takes these articles and produces one final summary. Start by dividing the documents into smaller sections and generating a summary for each section. Then pass those summaries to the LLM and ask it to identify the main points across them, remove repeated information, and write the final summary. If the documents are too large to process in one request, this approach lets you handle them in smaller batches.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">LangChain provides a map-reduce summarization workflow for this process, with code that you can adapt to your own collection of documents.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, an LLM, a collection of related documents, and LangChain.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/www.kaggle.com\/code\/ilhansevval\/summarize-documents-using-langchain\" target=\"_blank\" rel=\"noopener\">Summarize Documents with LangChain<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Multi-document summarization, document processing, context management, prompt design, LLM workflows.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>16. Structured Information Extraction Pipeline<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A resume might contain a person’s name, skills, education, and work history in completely different places. An invoice has its own set of fields, such as the invoice number, items, prices, and total amount. In both cases, the information is there, but it isn’t arranged in a format that another application can easily use.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Here, you can give an LLM a piece of unstructured text and define the fields you want it to return. For example, a receipt could be converted into fields such as order number, items, quantity, price, and total. Use a JSON Schema or Pydantic model to define that structure, then have the model extract the information and return it in that format. You can test the same approach on different types of documents and see how well the extracted fields match the source text.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Microsoft’s sample includes Python implementations for extracting structured information from text, PDFs, webpages, images, and GitHub content, so you can use one of these examples as the starting point for your pipeline.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, an LLM with structured-output support, Pydantic, and sample documents or text to process.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/learn.microsoft.com\/en-us\/azure\/developer\/ai\/how-to\/extract-entities-using-structured-outputs?tabs=github-codespaces\" target=\"_blank\" rel=\"noopener\">Entity Extraction with Azure OpenAI Structured Outputs – Microsoft Learn<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Structured information extraction, JSON outputs, schema design, Pydantic, document processing, prompt design.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>17. Prompt Evaluation Harness<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A prompt can give a good response for one question and a poor response for another. Changing a few words in the prompt can also change how the model responds. Testing it with a handful of examples isn’t enough if you want to know whether a new prompt is actually better.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Build a small test set with questions and expected answers, then run different versions of your prompt against the same examples. Record the responses and score them for things such as correctness, relevance, and consistency. You can then compare the results of each prompt version to avoid judging them one response at a time.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">LangSmith provides this workflow through datasets, evaluators, and experiments. Its evaluation tools let you run the same application against a dataset, score the outputs, and compare different experiments.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, an LLM API, LangSmith, a set of test inputs, and expected outputs or evaluation criteria.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/docs.langchain.com\/langsmith\/evaluation-quickstart\" target=\"_blank\" rel=\"noopener\">LangSmith Evaluation Quickstart<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> Prompt evaluation, automated testing, benchmarking, LLM evaluation, experiment comparison.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>18. LLM Guardrails & Moderation Layer<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">What happens when a user sends an unsafe request to your AI application? Or when the model generates a response that shouldn’t be shown to the user? A production application needs to check both sides of the interaction before allowing the request or response to continue.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Build a moderation layer that checks user prompts and model responses against a set of safety rules. You can start by detecting harmful content and setting thresholds for what should be allowed or blocked. Then add checks for issues such as prompt attacks and unsupported model responses. Microsoft Azure AI Content Safety provides APIs and Python examples for these checks, including severity-based filtering and Prompt Shields for detecting attacks in user inputs and documents.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Test the guardrails with safe and unsafe examples and see where the system blocks a request, allows it through, or flags it for further review.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>What you’ll need:<\/strong> Python, Azure AI Content Safety, an Azure AI resource, and sample prompts and model responses to test.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Source code\/tutorial:<\/strong> <a href=\"https:\/\/learn.microsoft.com\/en-us\/training\/modules\/moderate-content-detect-harm-azure-ai-content-safety\/\" target=\"_blank\" rel=\"noopener\">Implement generative AI guardrails with Azure AI Content Safety – Microsoft Learn<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Skills you’ll build:<\/strong> LLM guardrails, content moderation, prompt attack detection, safety filtering, responsible AI, safety pipelines.<\/p>\n\n\n<style data-wp-block-html=\"css\">\n<style>\n :root {\n --sp-brand: #255aad; \/* brand blue *\/\n --sp-brand-hover: #19468c;\n --sp-card-bg: #ffffff; \/* neutral-scale-0 *\/\n --sp-surface: #f4f6f9; \/* neutral-scale-2 *\/\n --sp-pill-bg: #f8fafc;\n --sp-border: #d6dee5; \/* neutral-scale-4 *\/\n --sp-text-primary: #1a2c47; \/* neutral-scale-12 *\/\n --sp-text-secondary: #61738e; \/* neutral-scale-9 *\/\n }\n\n .scaler-placement * { box-sizing: border-box; }\n\n .scaler-placement {\n font-family: \"Source Sans 3\", \"Source Sans Pro\", \"Roboto\", Arial, sans-serif;\n width: 100%;\n max-width: 100%;\n margin: 24px auto;\n display: flex;\n flex-direction: column;\n }\n\n \/* ---- Placement report (OnlineCallout) ---- *\/\n .scaler-placement__title {\n font-size: 24px;\n font-weight: 600;\n line-height: 1.3;\n color: var(--sp-text-primary);\n margin: 0 0 16px;\n }\n\n .scaler-placement__card {\n display: flex;\n flex-direction: column;\n width: 100%;\n background: var(--sp-card-bg);\n border: 1px solid var(--sp-border);\n border-radius: 20px;\n padding: 16px;\n box-shadow: 0 8px 32px rgba(37, 90, 173, 0.04);\n overflow: hidden;\n gap: 5px;\n }\n\n .scaler-placement__top {\n display: flex;\n flex-direction: row;\n align-items: center;\n width: 100%;\n gap: 18px;\n }\n\n .scaler-placement__stat-block {\n display: flex;\n flex-direction: column;\n justify-content: center;\n align-items: center;\n width: 30%;\n min-width: 200px;\n flex-shrink: 0;\n background: var(--sp-surface);\n border: 1px solid var(--sp-border);\n border-radius: 16px;\n padding: 16px 10px;\n gap: 8px;\n }\n .scaler-placement__stat-number {\n font-size: 48px;\n font-weight: 600;\n line-height: 1;\n color: var(--sp-brand);\n text-align: center;\n }\n .scaler-placement__stat-label {\n font-size: 13px;\n font-weight: 600;\n line-height: 14px;\n color: rgba(37, 90, 173, 0.85);\n text-transform: uppercase;\n letter-spacing: 0.8px;\n }\n\n .scaler-placement__info {\n display: flex;\n flex-direction: column;\n justify-content: center;\n flex: 1;\n gap: 10px;\n }\n .scaler-placement__proof-header {\n display: flex;\n flex-direction: row;\n align-items: center;\n gap: 8px;\n }\n .scaler-placement__proof-header svg { width: 18px; height: 18px; flex-shrink: 0; }\n .scaler-placement__proof-label {\n font-size: 12px;\n font-weight: 600;\n line-height: 16px;\n color: var(--sp-brand);\n text-transform: uppercase;\n letter-spacing: 1px;\n }\n .scaler-placement__body-text {\n font-size: 16px;\n font-weight: 400;\n line-height: 24px;\n color: var(--sp-text-primary);\n margin: 0;\n }\n\n .scaler-placement__bottom {\n display: flex;\n flex-direction: column;\n align-items: flex-start;\n margin-left: calc(30% + 18px);\n width: calc(70% - 18px);\n gap: 12px;\n }\n .scaler-placement__stats-row {\n display: flex;\n flex-direction: row;\n align-items: center;\n gap: 24px;\n }\n .scaler-placement__stat-item {\n display: flex;\n flex-direction: row;\n align-items: center;\n gap: 8px;\n }\n .scaler-placement__stat-item > svg {\n color: var(--sp-brand);\n width: 18px;\n height: 18px;\n flex-shrink: 0;\n }\n .scaler-placement__stat-text {\n display: flex;\n flex-direction: row;\n align-items: center;\n gap: 4px;\n }\n .scaler-placement__stat-value {\n font-size: 16px;\n font-weight: 600;\n line-height: 24px;\n color: var(--sp-text-primary);\n }\n .scaler-placement__stat-desc {\n font-size: 14px;\n font-weight: 400;\n line-height: 20px;\n color: var(--sp-text-secondary);\n }\n\n .scaler-placement__cta,\n .scaler-placement__cta:link,\n .scaler-placement__cta:visited {\n display: inline-flex;\n align-items: center;\n align-self: flex-start;\n font-size: 14px;\n font-weight: 600;\n line-height: 20px;\n color: var(--sp-brand) !important;\n gap: 6px;\n text-decoration: none;\n cursor: pointer;\n margin-top: 2px;\n transition: color 0.25s ease;\n }\n .scaler-placement__cta:hover,\n .scaler-placement__cta:focus { color: var(--sp-brand-hover) !important; }\n .scaler-placement__cta svg {\n width: 14px;\n height: 14px;\n fill: currentColor;\n transition: transform 0.25s ease;\n }\n .scaler-placement__cta:hover svg { transform: translateX(3px); }\n\n \/* ---- Hiring Partners ---- *\/\n .scaler-hiring {\n display: flex;\n align-items: center;\n justify-content: center;\n flex-wrap: wrap;\n gap: 8px 12px;\n margin: 24px 0 8px;\n padding-top: 16px;\n border-top: 1px solid var(--sp-border);\n }\n .scaler-hiring__label {\n font-size: 14px;\n color: var(--sp-text-secondary);\n white-space: nowrap;\n }\n .scaler-hiring__pills {\n display: flex;\n align-items: center;\n flex-wrap: wrap;\n gap: 8px;\n }\n .scaler-hiring__pill {\n display: inline-flex;\n align-items: center;\n gap: 6px;\n padding: 5px 12px 5px 5px;\n background: var(--sp-pill-bg);\n border-radius: 999px;\n }\n .scaler-hiring__badge {\n display: inline-flex;\n align-items: center;\n justify-content: center;\n flex-shrink: 0;\n width: 24px;\n height: 24px;\n border-radius: 50%;\n background: #ffffff;\n }\n .scaler-hiring__logo {\n width: 16px !important;\n height: 16px !important;\n object-fit: contain;\n margin: 0 !important;\n display: block;\n }\n .scaler-hiring__name {\n font-size: 14px;\n font-weight: 600;\n color: var(--sp-text-primary);\n }\n .scaler-hiring__more {\n font-size: 14px;\n color: var(--sp-text-secondary);\n }\n\n @media (max-width: 768px) {\n .scaler-placement__title { font-size: 18px; margin-bottom: 12px; }\n .scaler-placement__card { gap: 16px; border-radius: 16px; }\n .scaler-placement__top { gap: 12px; }\n .scaler-placement__stat-block { width: 35%; min-width: 110px; padding: 12px 8px; gap: 4px; }\n .scaler-placement__stat-number { font-size: 24px; }\n .scaler-placement__stat-label { font-size: 10px; }\n .scaler-placement__body-text { font-size: 14px; line-height: 20px; }\n .scaler-placement__bottom { margin-left: 0; width: 100%; gap: 16px; }\n .scaler-placement__stats-row { flex-wrap: wrap; gap: 10px 16px; }\n .scaler-placement__stat-value { font-size: 14px; line-height: 20px; }\n .scaler-placement__stat-desc { font-size: 12px; line-height: 16px; }\n }\n<\/style>\n\n<meta charset=\"UTF-8\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n\n <!-- Source Sans Pro (topics font) -->\n <link href=\"https:\/\/fonts.googleapis.com\/css2?family=Source+Sans+3:wght@400;600;700&display=swap\" rel=\"stylesheet\">\n\n<div class=\"scaler-placement\">\n <h2 class=\"scaler-placement__title\"><span class=\"ez-toc-section\" id=\"scaler-placement-report-and-statistics\"><\/span>Scaler Placement Report and Statistics<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n <div class=\"scaler-placement__card\">\n <div class=\"scaler-placement__top\">\n <div class=\"scaler-placement__stat-block\">\n <div class=\"scaler-placement__stat-number\">\u20b923L<\/div>\n <div class=\"scaler-placement__stat-label\">AVG CTC<\/div>\n <\/div>\n <div class=\"scaler-placement__info\">\n <div class=\"scaler-placement__proof-header\">\n <svg viewBox=\"0 0 21 21\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M8.64 0 0 8.683V21h12.257l8.64-8.683V0H8.64zm-.11 17.257H3.725v-4.828H8.53v4.828zm7.901-1.88-4.174 4.194V8.684H1.447l4.076-4.14h10.914l-.006 10.83v.004z\" fill=\"#255aad\"><\/path><\/svg>\n <span class=\"scaler-placement__proof-label\">SCALER PLACEMENT PROOF<\/span>\n <\/div>\n <p class=\"scaler-placement__body-text\">\n Scaler learners achieved 2.5x salary growth with average post-Scaler CTC reaching \u20b923L.\n <\/p>\n <\/div>\n <\/div>\n\n <div class=\"scaler-placement__bottom\">\n <div class=\"scaler-placement__stats-row\">\n <div class=\"scaler-placement__stat-item\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M11.199 7.957c-3.086 0-5.588 2.502-5.588 5.588s2.502 5.588 5.588 5.588c3.086 0 5.588-2.502 5.588-5.588s-2.502-5.588-5.588-5.588zM3.611 13.545c0-4.191 3.397-7.588 7.588-7.588s7.588 3.397 7.588 7.588c0 4.191-3.397 7.588-7.588 7.588s-7.588-3.397-7.588-7.588z\"><\/path><path d=\"M21.090 6.369h0.003c1.989 0 3.897 0.79 5.303 2.197s2.197 3.314 2.197 5.303-0.79 3.897-2.197 5.303c-1.406 1.407-3.314 2.197-5.303 2.197-0.552 0-1-0.448-1-1s0.448-1 1-1c1.459 0 2.858-0.579 3.889-1.611s1.611-2.43 1.611-3.889c0-1.459-0.58-2.858-1.611-3.889s-2.429-1.611-3.888-1.611c-0.507 0.002-1.013 0.070-1.502 0.203-0.533 0.145-1.082-0.17-1.227-0.703s0.17-1.082 0.703-1.227c0.659-0.179 1.339-0.27 2.021-0.272z\"><\/path><path d=\"M10.906 21.364c-1.605 0-3.186 0.386-4.61 1.126s-2.649 1.811-3.572 3.124c-0.318 0.452-0.941 0.561-1.393 0.243s-0.561-0.941-0.243-1.393c1.107-1.576 2.577-2.862 4.286-3.749s3.606-1.351 5.532-1.351 3.823 0.463 5.532 1.351c1.709 0.888 3.179 2.174 4.286 3.749 0.318 0.452 0.209 1.076-0.243 1.393s-1.076 0.209-1.393-0.243c-0.923-1.313-2.148-2.385-3.572-3.124s-3.005-1.126-4.61-1.126z\"><\/path><path d=\"M25.704 22.493c-1.424-0.74-3.005-1.125-4.61-1.124-0.552 0-1-0.447-1.001-0.999s0.447-1 0.999-1.001c1.926-0.001 3.824 0.462 5.533 1.349s3.179 2.174 4.286 3.751c0.317 0.452 0.208 1.076-0.244 1.393s-1.076 0.208-1.393-0.244c-0.922-1.313-2.147-2.385-3.571-3.125z\"><\/path><\/svg>\n <div class=\"scaler-placement__stat-text\">\n <span class=\"scaler-placement__stat-value\">11,000+<\/span>\n <span class=\"scaler-placement__stat-desc\">placements<\/span>\n <\/div>\n <\/div>\n <div class=\"scaler-placement__stat-item\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M1.2 27.5c0-0.442 0.358-0.8 0.8-0.8h28c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-28c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M3.727 4.227c0.338-0.338 0.795-0.527 1.273-0.527h12c0.477 0 0.935 0.19 1.273 0.527s0.527 0.795 0.527 1.273v22c0 0.442-0.358 0.8-0.8 0.8s-0.8-0.358-0.8-0.8v-22c0-0.053-0.021-0.104-0.059-0.141s-0.088-0.059-0.141-0.059h-12c-0.053 0-0.104 0.021-0.141 0.059s-0.059 0.088-0.059 0.141v22c0 0.442-0.358 0.8-0.8 0.8s-0.8-0.358-0.8-0.8v-22c0-0.477 0.19-0.935 0.527-1.273z\"><\/path><path d=\"M17.2 12.5c0-0.442 0.358-0.8 0.8-0.8h9c0.477 0 0.935 0.19 1.273 0.527s0.527 0.795 0.527 1.273v14c0 0.442-0.358 0.8-0.8 0.8s-0.8-0.358-0.8-0.8v-14c0-0.053-0.021-0.104-0.059-0.141s-0.088-0.059-0.141-0.059h-9c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M7.2 9.5c0-0.442 0.358-0.8 0.8-0.8h4c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-4c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M9.2 17.5c0-0.442 0.358-0.8 0.8-0.8h4c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-4c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M7.2 22.5c0-0.442 0.358-0.8 0.8-0.8h4c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-4c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M21.2 22.5c0-0.442 0.358-0.8 0.8-0.8h2c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-2c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><path d=\"M21.2 17.5c0-0.442 0.358-0.8 0.8-0.8h2c0.442 0 0.8 0.358 0.8 0.8s-0.358 0.8-0.8 0.8h-2c-0.442 0-0.8-0.358-0.8-0.8z\"><\/path><\/svg>\n <div class=\"scaler-placement__stat-text\">\n <span class=\"scaler-placement__stat-value\">650+<\/span>\n <span class=\"scaler-placement__stat-desc\">companies<\/span>\n <\/div>\n <\/div>\n <div class=\"scaler-placement__stat-item\">\n <svg viewBox=\"0 0 25 24\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M5.60938 18.8906C4.74688 18.0281 5.31875 16.2188 4.87813 15.1594C4.4375 14.1 2.75 13.1719 2.75 12C2.75 10.8281 4.41875 9.9375 4.87813 8.84063C5.3375 7.74375 4.74688 5.97187 5.60938 5.10938C6.47187 4.24688 8.28125 4.81875 9.34063 4.37813C10.4 3.9375 11.3281 2.25 12.5 2.25C13.6719 2.25 14.5625 3.91875 15.6594 4.37813C16.7562 4.8375 18.5281 4.24688 19.3906 5.10938C20.2531 5.97187 19.6812 7.78125 20.1219 8.84063C20.5625 9.9 22.25 10.8281 22.25 12C22.25 13.1719 20.5813 14.0625 20.1219 15.1594C19.6625 16.2562 20.2531 18.0281 19.3906 18.8906C18.5281 19.7531 16.7188 19.1812 15.6594 19.6219C14.6 20.0625 13.6719 21.75 12.5 21.75C11.3281 21.75 10.4375 20.0813 9.34063 19.6219C8.24375 19.1625 6.47187 19.7531 5.60938 18.8906Z\" fill=\"#ffffff\" stroke=\"#255aad\" stroke-width=\"1.4\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><\/path><path d=\"M16.625 9.75L11.1219 15L8.375 12.375\" stroke=\"#255aad\" stroke-width=\"1.4\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><\/path><\/svg>\n <span class=\"scaler-placement__stat-desc\">Verified data<\/span>\n <\/div>\n <\/div>\n\n <a class=\"scaler-placement__cta\" href=\"https:\/\/d2beiqkhq929f0.cloudfront.net\/public_assets\/assets\/000\/195\/869\/original\/Scaler_Career_Report.pdf?1778499863\" target=\"_blank\" rel=\"noopener\">\n See full placement report\n <svg viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"><\/path><\/svg>\n <\/a>\n <\/div>\n <\/div>\n\n <div class=\"scaler-hiring\">\n <span class=\"scaler-hiring__label\">Hiring Partners:<\/span>\n <div class=\"scaler-hiring__pills\">\n <span class=\"scaler-hiring__pill\">\n <span class=\"scaler-hiring__badge\"><img decoding=\"async\" class=\"scaler-hiring__logo\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/googleicon_.svg\" alt=\"Google\" loading=\"lazy\"><\/span>\n <span class=\"scaler-hiring__name\">Google<\/span>\n <\/span>\n <span class=\"scaler-hiring__pill\">\n <span class=\"scaler-hiring__badge\"><img decoding=\"async\" class=\"scaler-hiring__logo\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/amazonicon_.svg\" alt=\"Amazon\" loading=\"lazy\"><\/span>\n <span class=\"scaler-hiring__name\">Amazon<\/span>\n <\/span>\n <span class=\"scaler-hiring__pill\">\n <span class=\"scaler-hiring__badge\"><img decoding=\"async\" class=\"scaler-hiring__logo\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/microsoft.svg\" alt=\"Microsoft\" loading=\"lazy\"><\/span>\n <span class=\"scaler-hiring__name\">Microsoft<\/span>\n <\/span>\n <span class=\"scaler-hiring__pill\">\n <span class=\"scaler-hiring__badge\"><img decoding=\"async\" class=\"scaler-hiring__logo\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/flipkarticon_.svg\" alt=\"Flipkart\" loading=\"lazy\"><\/span>\n <span class=\"scaler-hiring__name\">Flipkart<\/span>\n <\/span>\n <span class=\"scaler-hiring__pill\">\n <span class=\"scaler-hiring__badge\"><img decoding=\"async\" class=\"scaler-hiring__logo\" src=\"https:\/\/d19zaobb0foqu6.cloudfront.net\/topics\/images\/adobeicon_.svg\" alt=\"Adobe\" loading=\"lazy\"><\/span>\n <span class=\"scaler-hiring__name\">Adobe<\/span>\n <\/span>\n <span class=\"scaler-hiring__more\">1200+ more<\/span>\n <\/div>\n <\/div>\n <\/div>\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"datasets-evaluation-the-metrics-that-matter\"><\/span><strong>Datasets, Evaluation & the Metrics That Matter<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Every NLP model is shaped by the data it's trained and tested on. A sentiment classifier needs labelled opinions, a summarization model requires reference summaries, while a RAG system depends on the quality of the documents it retrieves. Choosing the right dataset is the first step in building a reliable NLP application; evaluating it with the right metrics is what tells you whether it's actually performing as expected.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Where to Find NLP Datasets<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">You don't always need to collect or annotate your own data. Many NLP datasets are publicly available and cover tasks ranging from text classification to question answering and summarization.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/huggingface.co\/datasets\" target=\"_blank\" rel=\"noopener\"><strong>Hugging Face Datasets<\/strong><\/a> - Thousands of datasets for classification, summarization, translation, question answering, and more.<\/li>\n\n\n\n<li><a href=\"https:\/\/www.kaggle.com\/datasets?tags=13204-NLP\" target=\"_blank\" rel=\"noopener\"><strong>Kaggle<\/strong><\/a> - Community-contributed datasets and NLP competitions.<\/li>\n\n\n\n<li><a href=\"https:\/\/archive.ics.uci.edu\/datasets\" target=\"_blank\" rel=\"noopener\"><strong>UCI Machine Learning Repository<\/strong> <\/a>- Classic datasets for machine learning and NLP experiments.<\/li>\n\n\n\n<li><a href=\"https:\/\/commoncrawl.org\/\" target=\"_blank\" rel=\"noopener\"><strong>Common Crawl & Wikipedia Dumps<\/strong><\/a> - Large-scale text corpora for language modelling and information retrieval.<\/li>\n\n\n\n<li><a href=\"https:\/\/data.gov.in\/\" target=\"_blank\" rel=\"noopener\"><strong>Government open-data portals & research repositories<\/strong><\/a> - Useful for domain-specific projects in healthcare, finance, legal tech, and public policy.<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Evaluation Metrics by NLP Task<\/strong><\/h3>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><tbody><tr><td><strong>NLP Task<\/strong><\/td><td><strong>Common Metrics<\/strong><\/td><td><strong>What They Do<\/strong><\/td><\/tr><tr><td>Text Classification (Spam, Sentiment, Emotion Detection)<\/td><td>Accuracy, Precision, Recall, F1-score<\/td><td>Measures how well the model distinguishes between different classes while balancing false positives and false negatives.<\/td><\/tr><tr><td>Named Entity Recognition (NER)<\/td><td>Precision, Recall, F1-score<\/td><td>Evaluates how accurately entities are identified and extracted from text.<\/td><\/tr><tr><td>Semantic Search & Retrieval<\/td><td>Precision@K, Recall@K, MRR, nDCG<\/td><td>Measures whether the most relevant documents are retrieved and ranked highly.<\/td><\/tr><tr><td>Text Summarization<\/td><td>ROUGE, BLEU, Human Evaluation<\/td><td>Compares generated summaries against reference summaries while assessing readability and information coverage.<\/td><\/tr><tr><td>Question Answering<\/td><td>Exact Match (EM), F1-score<\/td><td>Evaluates whether the predicted answer matches the expected response and captures the correct context.<\/td><\/tr><tr><td>RAG Applications<\/td><td>Retrieval Accuracy, Faithfulness, Context Precision, Human Evaluation<\/td><td>Assesses both retrieval quality and whether responses remain grounded in the retrieved documents.<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Beyond Benchmark Scores<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Evaluation doesn't end with a benchmark score, especially for modern LLM applications. Several models may achieve similar metrics while producing noticeably different user experiences. That's why many production systems combine automated evaluation with human review to assess factors such as factual accuracy, relevance, completeness, and consistency.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">When showcasing an nlp project, include not only the model and results but also the dataset you used, the evaluation metrics you tracked, and why those metrics were appropriate for the task. It demonstrates that you've thought beyond implementation and understand how NLP systems are assessed.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"choosing-projects-for-final-year-vs-job-portfolio\"><\/span><strong>Choosing Projects for Final Year vs Job Portfolio<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The project that works well for a college submission may not be the one you want at the top of your resume. An nlp projects for final year needs enough depth for you to spend time on the problem and defend your work. A portfolio needs projects that show different areas of NLP.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>If You're Choosing a Final-Year Project<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">With a Resume Parser, you can work with resumes that use different formats and layouts, compare how well your parser extracts skills, education, and experience, and see where it starts making mistakes.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">With a News Search Engine, you can test different search queries, compare how the results are ranked, and check whether the articles appearing at the top are actually relevant.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The same idea works for a Question Answering System or Multi-Document Summarizer. Pick a problem where there is something to test and improve after the first version works. That gives you material for your report and viva without forcing several unrelated techniques into one project.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>If You're Building a Portfolio for Interviews<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A portfolio is easier to judge when the projects don't all look the same.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You might start with a Spam Classifier using TF-IDF, then move to BERT Fine-Tuning for a transformer-based task. Semantic Search introduces embeddings and vector search, while RAG adds document retrieval and LLMs. Prompt Evaluation then covers a different part of the LLM workflow: checking whether the model is actually giving good responses.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You don't need every one of these projects. Pick the ones that fit the roles you're applying for and avoid filling the portfolio with five versions of text classification.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>What Should You Know About a Project Before Adding It to Your Resume?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">If your project uses TF-IDF, you should know how the text is converted into features and why you chose it. If you fine-tuned BERT, you should understand tokenization, attention, and what fine-tuning changes. If you built a RAG chatbot, you should know how the documents are split, how the relevant sections are retrieved, and what happens when the retrieved information is wrong.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The same applies to the other projects in this list. You don't have to know every library function by memory. You should understand the actual process you built well enough to explain it without going back to the tutorial.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This is the level of preparation that helps when the project moves from a GitHub repository to a viva or technical interview.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"nlp-vs-llm-engineering-understanding-the-difference\"><\/span><strong>NLP vs LLM Engineering: Understanding the Difference<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">You have now worked through projects that range from spam classification and NER to RAG and LLM guardrails. But where do these projects fit when you start looking at NLP and LLM engineering jobs?<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">NLP Engineer and LLM Engineer are not completely separate roles. There is a lot of overlap, but the day-to-day work can be quite different.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">An NLP Engineer may work on text classification, sentiment analysis, NER, document retrieval, or transformer models. An LLM Engineer may work on semantic search, RAG, structured extraction, prompt evaluation, or AI safety.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><tbody><tr><td><strong>Career Path<\/strong><\/td><td><strong>Common Responsibilities<\/strong><\/td><td><strong>Projects from This Guide<\/strong><\/td><\/tr><tr><td><strong>NLP Engineer<\/strong><\/td><td>Text classification, sentiment analysis, NER, document retrieval, transformer fine-tuning<\/td><td>Spam Email Classifier, Twitter Sentiment Analysis, Resume Parser, News Search Engine, BERT Sentiment Analysis<\/td><\/tr><tr><td><strong>LLM Engineer<\/strong><\/td><td>Semantic search, RAG, prompt engineering, structured extraction, evaluation, AI safety<\/td><td>Semantic Search, RAG Chatbot, Multi-Document Summarizer, Prompt Evaluation Harness, LLM Guardrails<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">You don't have to decide which title fits you right now. The projects themselves can help you figure that out. If you enjoy working with datasets, features, classification, extraction, and model performance, the earlier NLP projects give you more of that experience. If you prefer building applications around language models, the later projects move into retrieval, generation, evaluation, and safety.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">There is also a reason not to skip the earlier projects just because LLMs are getting more attention. A RAG system still needs retrieval. Semantic search still needs embeddings. Any model still needs evaluation. The later projects use many of the ideas you encountered earlier, but in larger applications.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">So you don't need to choose between \u201cNLP\u201d and \u201cLLM\u201d before you've had a chance to try both. Build across the two, see which work interests you, and then use that experience to decide what you want your portfolio to focus on.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">If you want structured learning alongside these projects, you can explore <a href=\"https:\/\/www.scaler.com\/ai-machine-learning-course\/\">Scaler's AI & ML Program<\/a>.<\/p>\n\n\n<meta charset=\"UTF-8\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n\n <!-- Source Sans Pro (topics font) -->\n <link href=\"https:\/\/fonts.googleapis.com\/css2?family=Source+Sans+3:wght@400;600;700&display=swap\" rel=\"stylesheet\">\n <link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.css\">\n <script src=\"https:\/\/cdn.jsdelivr.net\/npm\/swiper@11\/swiper-bundle.min.js\"><\/script>\n\n <style>\n :root {\n --ss-primary: #005bb3; \/* primary-color-80 *\/\n --ss-card-bg: #ffffff;\n --ss-border: #e6e6e6;\n --ss-quote-bg: #f6f6f6;\n --ss-text-dark: #1c1c1c;\n --ss-text-mid: #6b6b6b;\n --ss-text-muted: #9a9a9a;\n --ss-radius: 12px;\n --ss-radius-sm: 8px;\n --ss-shadow: 0 8px 16px 0 rgba(50, 71, 102, 0.16);\n }\n\n .scaler-ss * { box-sizing: border-box; }\n\n .scaler-ss {\n font-family: \"Source Sans 3\", \"Source Sans Pro\", \"Roboto\", Arial, sans-serif;\n padding: 24px;\n width: 100%;\n }\n .scaler-ss__inner {\n max-width: 1000px;\n margin: 0 auto;\n position: relative;\n width: 100%;\n }\n\n \/* Heading *\/\n .scaler-ss__heading {\n font-size: 26px;\n line-height: 34px;\n font-weight: 700;\n margin: 0 0 16px;\n text-align: center;\n color: var(--ss-text-dark);\n }\n\n \/* Swiper *\/\n .scaler-ss__swiper-wrap { position: relative; }\n .scaler-ss .swiper { width: 100%; padding: 4px; }\n .scaler-ss-swiper .swiper-slide {\n width: 300px; \/* topics card width *\/\n height: auto;\n display: flex;\n }\n\n \/* Card *\/\n .scaler-ss-card {\n display: flex;\n flex-direction: column;\n width: 100%;\n border: 1px solid var(--ss-border);\n border-radius: var(--ss-radius);\n overflow: hidden;\n background: var(--ss-card-bg);\n }\n\n .scaler-ss-card__hike {\n padding: 8px 12px;\n background: var(--ss-primary);\n color: #ffffff;\n font-size: 14px;\n font-weight: 700;\n text-align: center;\n letter-spacing: 0.4px;\n }\n\n .scaler-ss-card__body {\n display: flex;\n flex: 1 1 auto;\n flex-direction: column;\n gap: 12px;\n padding: 16px;\n }\n\n \/* Person *\/\n .scaler-ss-card__person {\n display: flex;\n align-items: center;\n gap: 10px;\n }\n .scaler-ss-card__avatar {\n width: 44px;\n height: 44px;\n min-width: 44px;\n border-radius: 50%;\n object-fit: cover;\n background: var(--ss-quote-bg);\n }\n .scaler-ss-card__name {\n font-size: 16px;\n font-weight: 700;\n line-height: 20px;\n color: var(--ss-text-dark);\n }\n .scaler-ss-card__city {\n font-size: 13px;\n line-height: 18px;\n color: var(--ss-text-mid);\n }\n\n \/* Before \/ Now transition *\/\n .scaler-ss-card__transition {\n display: flex;\n align-items: center;\n justify-content: space-between;\n gap: 8px;\n }\n .scaler-ss-card__col {\n flex: 1;\n min-width: 0;\n }\n .scaler-ss-card__col-label {\n font-size: 11px;\n font-weight: 600;\n letter-spacing: 0.4px;\n color: var(--ss-text-muted);\n margin-bottom: 2px;\n }\n .scaler-ss-card__col-role {\n font-size: 13px;\n line-height: 18px;\n color: var(--ss-text-mid);\n }\n .scaler-ss-card__col-company {\n font-size: 14px;\n font-weight: 700;\n line-height: 20px;\n color: var(--ss-text-dark);\n }\n .scaler-ss-card__company-logo {\n margin-top: 2px;\n max-width: 64px;\n max-height: 20px;\n width: auto;\n height: auto;\n object-fit: contain;\n display: block;\n }\n .scaler-ss-card__arrow {\n flex-shrink: 0;\n width: 28px;\n height: 8px;\n color: var(--ss-text-mid);\n fill: currentColor;\n }\n\n \/* Quote *\/\n .scaler-ss-card__quote {\n font-size: 13px;\n font-style: italic;\n line-height: 19px;\n color: var(--ss-text-mid);\n background: var(--ss-quote-bg);\n border-radius: var(--ss-radius-sm);\n padding: 8px 10px;\n }\n\n \/* View Profile *\/\n .scaler-ss-card__cta,\n .scaler-ss-card__cta:link,\n .scaler-ss-card__cta:visited {\n display: block;\n margin-top: auto;\n text-align: center;\n font-size: 14px;\n font-weight: 600;\n color: var(--ss-primary) !important;\n text-decoration: none;\n border: 1px solid var(--ss-primary);\n border-radius: var(--ss-radius-sm);\n padding: 9px 12px;\n cursor: pointer;\n transition: background 0.2s ease, color 0.2s ease;\n }\n .scaler-ss-card__cta:hover,\n .scaler-ss-card__cta:focus {\n background: var(--ss-primary);\n color: #ffffff !important;\n }\n\n \/* Side navigation arrows *\/\n .scaler-ss__nav {\n position: absolute !important;\n top: 50% !important;\n transform: translateY(-50%) !important;\n z-index: 2;\n width: 46px !important;\n height: 46px !important;\n min-width: 46px;\n display: flex !important;\n align-items: center;\n justify-content: center;\n padding: 0 !important;\n background: var(--ss-card-bg) !important;\n border: none !important;\n border-radius: var(--ss-radius-sm) !important;\n box-shadow: var(--ss-shadow) !important;\n color: #61738e !important;\n cursor: pointer;\n transition: opacity 0.2s ease, visibility 0.2s ease;\n }\n .scaler-ss__nav--prev { left: 8px !important; }\n .scaler-ss__nav--next { right: 8px !important; }\n .scaler-ss__nav svg {\n width: 20px !important;\n height: 20px !important;\n fill: #61738e !important;\n }\n .scaler-ss__nav.swiper-button-disabled {\n opacity: 0;\n visibility: hidden;\n pointer-events: none;\n }\n\n .scaler-ss .swiper-button-prev,\n .scaler-ss .swiper-button-next,\n .scaler-ss .swiper-pagination { display: none; }\n\n \/* Skeleton *\/\n .scaler-ss-card--skeleton { pointer-events: none; }\n .scaler-ss-card--skeleton .scaler-ss-card__hike,\n .scaler-ss-card--skeleton .scaler-ss-card__avatar,\n .scaler-ss-card--skeleton .scaler-ss-card__person-text,\n .scaler-ss-card--skeleton .scaler-ss-card__transition,\n .scaler-ss-card--skeleton .scaler-ss-card__quote,\n .scaler-ss-card--skeleton .scaler-ss-card__cta {\n background: linear-gradient(90deg, var(--ss-border) 25%, var(--ss-quote-bg) 50%, var(--ss-border) 75%);\n background-size: 200% 100%;\n animation: ss-shimmer 1.5s infinite;\n color: transparent !important;\n border-color: transparent !important;\n border-radius: 4px;\n }\n .scaler-ss-card--skeleton .scaler-ss-card__person-text { flex: 1; height: 32px; }\n .scaler-ss-card--skeleton .scaler-ss-card__transition { height: 56px; }\n .scaler-ss-card--skeleton .scaler-ss-card__quote { height: 48px; }\n .scaler-ss-card--skeleton .scaler-ss-card__cta { height: 38px; }\n\n @keyframes ss-shimmer {\n 0% { background-position: 200% 0; }\n 100% { background-position: -200% 0; }\n }\n\n @media (max-width: 768px) {\n .scaler-ss { padding: 16px; }\n .scaler-ss__heading { font-size: 22px; line-height: 30px; }\n .scaler-ss-swiper .swiper-slide { width: 270px; }\n .scaler-ss__nav { display: none !important; }\n }\n <\/style>\n\n\n\n <div class=\"scaler-ss js-scaler-ss\">\n\n <template class=\"js-ss-card-template\">\n <div class=\"swiper-slide\">\n <article class=\"scaler-ss-card\">\n <div class=\"scaler-ss-card__hike js-ss-hike\"><\/div>\n <div class=\"scaler-ss-card__body\">\n <div class=\"scaler-ss-card__person\">\n <img decoding=\"async\" class=\"scaler-ss-card__avatar js-ss-avatar\" src=\"\" alt=\"\" loading=\"lazy\">\n <div class=\"scaler-ss-card__person-text\">\n <div class=\"scaler-ss-card__name js-ss-name\"><\/div>\n <div class=\"scaler-ss-card__city js-ss-city\"><\/div>\n <\/div>\n <\/div>\n\n <div class=\"scaler-ss-card__transition\">\n <div class=\"scaler-ss-card__col\">\n <div class=\"scaler-ss-card__col-label\">BEFORE:<\/div>\n <div class=\"scaler-ss-card__col-role js-ss-before-role\"><\/div>\n <div class=\"scaler-ss-card__col-company js-ss-before-company\"><\/div>\n <\/div>\n <svg class=\"scaler-ss-card__arrow\" viewBox=\"0 0 32 8\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M31.3882 4.35355C31.5835 4.15829 31.5835 3.84171 31.3882 3.64645L28.2062 0.464466C28.011 0.269204 27.6944 0.269204 27.4991 0.464466C27.3039 0.659728 27.3039 0.976311 27.4991 1.17157L30.3276 4L27.4991 6.82843C27.3039 7.02369 27.3039 7.34027 27.4991 7.53553C27.6944 7.7308 28.011 7.7308 28.2062 7.53553L31.3882 4.35355ZM0.034668 4.5H31.0347V3.5H0.034668V4.5Z\"><\/path><\/svg>\n <div class=\"scaler-ss-card__col\">\n <div class=\"scaler-ss-card__col-label\">NOW:<\/div>\n <div class=\"scaler-ss-card__col-role js-ss-now-role\"><\/div>\n <div class=\"js-ss-now-company\"><\/div>\n <\/div>\n <\/div>\n\n <div class=\"scaler-ss-card__quote js-ss-quote\"><\/div>\n\n <a class=\"scaler-ss-card__cta js-ss-cta\" href=\"#\" target=\"_blank\" rel=\"noopener\">View Profile<\/a>\n <\/div>\n <\/article>\n <\/div>\n <\/template>\n\n <div class=\"scaler-ss__inner\">\n <h2 class=\"scaler-ss__heading\"><span class=\"ez-toc-section\" id=\"scaler-alumni-and-their-success-stories\"><\/span>Scaler Alumni and Their Success Stories<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n <div class=\"scaler-ss__swiper-wrap\">\n <div class=\"swiper scaler-ss-swiper\">\n <div class=\"swiper-wrapper js-ss-wrapper\"><\/div>\n <\/div>\n\n <button class=\"scaler-ss__nav scaler-ss__nav--prev js-ss-prev\" aria-label=\"Previous\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(-90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"><\/path><\/svg>\n <\/button>\n <button class=\"scaler-ss__nav scaler-ss__nav--next js-ss-next\" aria-label=\"Next\">\n <svg fill=\"currentColor\" viewBox=\"0 0 32 32\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" style=\"transform: rotate(90deg);\"><path d=\"M5.313 16l10.688-10.688 10.688 10.688-1.938 1.875-7.438-7.438v16.25h-2.625v-16.25l-7.5 7.438z\"><\/path><\/svg>\n <\/button>\n <\/div>\n <\/div>\n <\/div>\n\n <script>\n document.addEventListener(\"DOMContentLoaded\", () => {\n\n const ASSET_CDN = \"https:\/\/d2beiqkhq929f0.cloudfront.net\/public_assets\/assets\";\n const AVATAR_FALLBACK =\n \"data:image\/svg+xml,%3Csvg xmlns='http:\/\/www.w3.org\/2000\/svg' width='44' height='44' viewBox='0 0 44 44'%3E%3Ccircle cx='22' cy='22' r='22' fill='%23e6e6e6'\/%3E%3Ccircle cx='22' cy='17' r='7' fill='%23b9b9b9'\/%3E%3Cpath d='M8 40c0-8 6.3-13 14-13s14 5 14 13' fill='%23b9b9b9'\/%3E%3C\/svg%3E\";\n\n \/\/ Mirrors buildAssetUrl(): pass through absolute urls, else build the CDN url.\n function buildAssetUrl(raw) {\n if (!raw) return \"\";\n if (String(raw).startsWith(\"http\")) return raw;\n return `${ASSET_CDN}\/${raw}\/original`;\n }\n\n const instances = document.querySelectorAll(\".js-scaler-ss\");\n\n instances.forEach(container => {\n if (container.dataset.initialized === \"true\") return;\n container.dataset.initialized = \"true\";\n\n const swiperElement = container.querySelector(\".scaler-ss-swiper\");\n const wrapper = container.querySelector(\".js-ss-wrapper\");\n const template = container.querySelector(\".js-ss-card-template\");\n const nextBtn = container.querySelector(\".js-ss-next\");\n const prevBtn = container.querySelector(\".js-ss-prev\");\n\n if (!wrapper || !template) {\n console.error(\"Scaler Success Stories: missing required elements\");\n return;\n }\n\n const swiper = new Swiper(swiperElement, {\n slidesPerView: \"auto\",\n spaceBetween: 16,\n grabCursor: true,\n observer: true,\n observeParents: true,\n navigation: { nextEl: nextBtn, prevEl: prevBtn },\n });\n\n const pathParts = location.pathname.split(\"\/\").filter(Boolean);\n const currentSlug = pathParts.length > 0 ? pathParts[pathParts.length - 1] : \"homepage\";\n const profileUrl =\n `https:\/\/www.scaler.com\/review\/?utm_source=blog&utm_medium=success_stories&utm_content=${currentSlug}`;\n\n function showSkeletons(count = 4) {\n wrapper.innerHTML = \"\";\n for (let i = 0; i < count; i++) {\n const clone = template.content.cloneNode(true);\n clone.querySelector(\".scaler-ss-card\").classList.add(\"scaler-ss-card--skeleton\");\n wrapper.appendChild(clone);\n }\n swiper.update();\n }\n\n function renderStories(stories) {\n wrapper.innerHTML = \"\";\n if (!stories.length) {\n container.style.display = \"none\"; \/\/ match topics: render nothing when empty\n return;\n }\n\n stories.forEach(story => {\n const a = story.attributes || {};\n const clone = template.content.cloneNode(true);\n\n const hike = String(a.growth || \"\").replace(\"%\", \"\").trim() || \"150\";\n clone.querySelector(\".js-ss-hike\").textContent = `${hike}% HIKE`;\n\n \/\/ Avatar\n const avatar = buildAssetUrl(a.image_url) || a.generated_thumbnail_url || \"\";\n const img = clone.querySelector(\".js-ss-avatar\");\n img.src = avatar || AVATAR_FALLBACK;\n img.alt = a.name || \"Scaler alumnus\";\n img.onerror = function () { this.onerror = null; this.src = AVATAR_FALLBACK; };\n\n clone.querySelector(\".js-ss-name\").textContent = a.name || \"\";\n const cityEl = clone.querySelector(\".js-ss-city\");\n if (a.city) cityEl.textContent = a.city; else cityEl.style.display = \"none\";\n\n \/\/ BEFORE\n clone.querySelector(\".js-ss-before-role\").textContent = a.pre_scaler_role || a.pre_academy_company || \"\";\n const beforeCompany = clone.querySelector(\".js-ss-before-company\");\n if (a.pre_academy_company) beforeCompany.textContent = a.pre_academy_company;\n else beforeCompany.style.display = \"none\";\n\n \/\/ NOW\n clone.querySelector(\".js-ss-now-role\").textContent = a.role || \"\";\n const nowCompanyBox = clone.querySelector(\".js-ss-now-company\");\n const companyLogo = buildAssetUrl(a.company_logo_url);\n if (companyLogo) {\n const logo = document.createElement(\"img\");\n logo.className = \"scaler-ss-card__company-logo\";\n logo.src = companyLogo;\n logo.alt = a.post_academy_company || \"company\";\n logo.loading = \"lazy\";\n logo.onerror = function () {\n this.onerror = null;\n this.replaceWith(Object.assign(document.createElement(\"div\"), {\n className: \"scaler-ss-card__col-company\",\n textContent: a.post_academy_company || \"\",\n }));\n };\n nowCompanyBox.appendChild(logo);\n } else if (a.post_academy_company) {\n nowCompanyBox.className = \"scaler-ss-card__col-company\";\n nowCompanyBox.textContent = a.post_academy_company;\n }\n\n \/\/ Quote (clip to 110 chars like topics)\n const quoteEl = clone.querySelector(\".js-ss-quote\");\n quoteEl.textContent = `\u201c${String(a.content).slice(0, 110)}\u2026\u201d`;\n\n \/\/ View Profile\n clone.querySelector(\".js-ss-cta\").href = profileUrl;\n\n wrapper.appendChild(clone);\n });\n\n swiper.update();\n swiper.slideTo(0);\n }\n\n async function fetchStories() {\n try {\n showSkeletons();\n \/\/ Same-origin POST on www.scaler.com (the blog canonical host).\n \/\/ The endpoint only returns JSON when the Referer is a \/topics\/ page\n \/\/ (a soft anti-hotlinking gate); a \/blog\/ referer gets a Turbolinks\n \/\/ redirect. Since blog + topics are the same origin, we set the\n \/\/ fetch referrer to a topics URL so the server serves the data.\n const res = await fetch(\"\/success-stories\/testimonials\", {\n method: \"POST\",\n credentials: \"include\",\n referrer: \"https:\/\/www.scaler.com\/topics\/\",\n referrerPolicy: \"unsafe-url\",\n headers: {\n \"Content-Type\": \"application\/json\",\n \"Accept\": \"application\/json\",\n \"X-Requested-With\": \"XMLHttpRequest\",\n },\n \/\/ q:{} => all programs (no program filter)\n body: JSON.stringify({\n page_number: 1,\n per_page: 12,\n sort_by: \"eligible_for_qna\",\n new_quick_filters: true,\n quick_filter: [],\n q: {},\n }),\n });\n const json = await res.json();\n const all = (json && json.data && json.data.data) || [];\n \/\/ Only show stories that carry a testimonial quote (matches topics).\n const withQuote = all.filter(s => s.attributes && s.attributes.content);\n renderStories(withQuote);\n } catch (error) {\n console.error(\"Failed to load success stories:\", error);\n container.style.display = \"none\";\n }\n }\n\n fetchStories();\n });\n });\n <\/script>\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"faqs\"><\/span><strong>FAQs<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q1. What are good NLP projects for beginners?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">If you're new to NLP, start with projects that teach the fundamentals of text processing before moving on to transformer models. A spam email classifier using TF-IDF, Twitter sentiment analysis, and a Named Entity Recognition (NER) pipeline with spaCy are some good options to begin with because they introduce preprocessing, feature engineering, and text classification while remaining relevant for technical interviews.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q2. Is classical NLP still worth learning in the LLM era?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Yes. Classical NLP techniques such as TF-IDF, text classification, and Named Entity Recognition are still widely used in production systems where speed, interpretability, and lower computational costs matter. They also provide the foundation for understanding transformer models and remain common topics in NLP and machine learning interviews.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q3. Which NLP projects stand out in 2026 interviews?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Projects that solve real-world problems tend to make the strongest impression. A Retrieval-Augmented Generation (RAG) chatbot built on your own documents, a semantic search engine using embeddings, or an LLM evaluation framework demonstrate practical experience with modern NLP workflows while giving you meaningful architectural decisions to discuss during interviews.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q4. Do NLP projects require a GPU?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Not always. Classical NLP projects such as text classification, sentiment analysis, and Named Entity Recognition can run comfortably on a standard laptop. Many transformer-based projects can be trained using free GPU resources on platforms like Google Colab, while several LLM applications rely on hosted APIs and therefore don't require a dedicated GPU at all.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q5. How are NLP projects evaluated?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The evaluation metric depends on the task. Classification projects typically use Accuracy, Precision, Recall, and F1-score, while summarization models are commonly evaluated using ROUGE. Retrieval-based applications such as semantic search or RAG systems are measured using retrieval metrics like Precision@K or Recall@K, often supplemented with human evaluation to assess response quality and factual accuracy.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Q6. Which datasets are commonly used for NLP projects?<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Popular NLP datasets include IMDb Reviews for sentiment analysis, TweetEval for social media classification, SQuAD for question answering, and CNN\/DailyMail for text summarization. If you're building a RAG application, you can also use your own PDFs, documentation, research papers, or knowledge base as the source documents.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>University NLP projects are already working with the tools behind today\u2019s language applications. Stanford\u2019s CS224N course still covers the basics of NLP. Still, its 2026 coursework also includes BERT, pretraining, LoRA, and other parameter-efficient fine-tuning methods, prompting, RAG, tool use, and language-model evaluation. Students finish the course with a research project where they test their […]<\/p>\n","protected":false},"author":242,"featured_media":14250,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[316,37,328,410],"tags":[560],"class_list":["post-14248","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","category-artificial-intelligence-machine-learning","category-nlp","category-projects","tag-nlp-project-ideas"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/posts\/14248","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/users\/242"}],"replies":[{"embeddable":true,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/comments?post=14248"}],"version-history":[{"count":1,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/posts\/14248\/revisions"}],"predecessor-version":[{"id":14251,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/posts\/14248\/revisions\/14251"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/media\/14250"}],"wp:attachment":[{"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/media?parent=14248"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/categories?post=14248"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.scaler.com\/blog\/wp-json\/wp\/v2\/tags?post=14248"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}