Nykaa is hiring a RAG & LLM Engineer Intern in Ghaziabad — a hands-on, mentor-led internship built for students and 2026 freshers who want real exposure to retrieval-augmented generation rather than a certificate-only programme. About the role As a RAG & LLM Engineer Intern at Nykaa, your core focus will be building retrieval pipelines that keep model answers grounded in real documents. You are attached to a named mentor, join the team's stand-ups and reviews, and are trusted with work that matters rather than filler tasks. Questions are encouraged, your output is reviewed properly, and you get the feedback you need to actually improve. What you'll do day to day • Work on building retrieval pipelines that keep model answers grounded in real documents alongside the core team • Use Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings, Python, LangChain / LlamaIndex on live, production-facing problems • Turn your work into clear documentation, notes or decks for reviewers • Take part in reviews and act on the feedback you receive • Collaborate across functions — engineering, product, design or business as the work demands • Follow the team's quality, testing and documentation standards • Track and report your own progress against an agreed plan Skills you'll build You will build practical, demonstrable experience in Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings, Python, LangChain / LlamaIndex, Chunking Strategies, LLM Evaluation, Document Parsing, REST APIs, Latency & Cost Tuning — the exact toolkit that separates a hireable fresher from a CV full of coursework. No prior work experience is required. Curiosity, problem-solving ability and a willingness to learn matter far more here than a polished résumé, and every one of these skills is taught on the job. Who should apply Students in their final year or recent graduates from B.E./B.Tech (CSE/IT/AI-ML), MCA or M.Sc (CS / Data Science) based in or willing to work from Ghaziabad. You should be willing to learn Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings and able to communicate clearly in writing. Personal projects, coursework, certifications or open-source contributions strengthen your application, but none of them are mandatory. Internship details This is a 3-month internship in Ghaziabad, Uttar Pradesh, with a stipend of ₹23,000 per month. You will be given a defined project so that you leave with a RAG pipeline whose answers you measured against ground truth. Working hours follow the team's normal business hours and the internship includes a certificate on successful completion. You will finish this internship with practical, demonstrable skills, real project work to point at, and a much clearer sense of how the industry actually operates day to day. Strong performers are considered for a Pre-Placement Offer (PPO) and a full-time role. Eligibility criteria • Education: B.E./B.Tech (CSE/IT/AI-ML), MCA or M.Sc (CS / Data Science) • Graduating batch: 2026, 2027 or 2028 — final-year students and recent graduates are both eligible • Academics: 60% / 6.0 CGPA and above preferred; consistent academics help, but a strong project or portfolio is weighed alongside marks • Technical skills: Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings, Python, LangChain / LlamaIndex • Also useful: Chunking Strategies, LLM Evaluation, Document Parsing, REST APIs, Latency & Cost Tuning • Soft skills: clear written and spoken English, ownership of your work, comfort asking questions, and the discipline to check your own work before you ship it • Experience: none required — this is a fresher internship. Academic projects, certifications, GitHub work or a previous internship strengthen an application but are not mandatory • Location: based in Ghaziabad, Uttar Pradesh or willing to work from there for the internship period Additional information • Joining date: immediate / rolling — selected candidates usually start within 2-4 weeks of accepting the offer, and start dates can be adjusted around end-semester exams • Working hours: Monday to Friday, standard business hours (about 8-9 hours a day) at the office • Duration: 3 months, with a certificate issued on successful completion • Documents at joining: college ID or degree certificate, a government photo ID, and a No-Objection Certificate from your college where your programme requires one • Equipment: a workstation, software licences and all tool access are provided at the office — you do not need to bring your own machine • Bond: no bond and no service agreement is attached to this internship • Fees: Nykaa and MyInternships.in never charge a candidate any fee at any stage of the application, selection or onboarding process. Report anyone who asks you for money in the name of this internship How to apply Click Apply Now. If you are not logged in you will be asked to sign in and register first, after which you will be taken to Nykaa's official careers page to complete your application. Applications are reviewed on a rolling basis and seats are limited, so apply early. Responsibilities: Work on building retrieval pipelines that keep model answers grounded in real documents alongside the core team; Apply Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings, Python to live, production-facing problems; Document your work clearly for reviewers and stakeholders; Participate in reviews and act on the feedback you receive; Collaborate across engineering, product, design or business as required; Follow the team's quality, testing and documentation standards. Requirements: Pursuing or recently completed B.E./B.Tech (CSE/IT/AI-ML), MCA or M.Sc (CS / Data Science); Graduating batch 2026, 2027 or 2028 — final-year students and recent graduates welcome; 60% / 6.0 CGPA and above preferred; strong project work is considered alongside academics; Working knowledge of Retrieval-Augmented Generation (RAG), Vector Databases (Pinecone / Chroma), Embeddings (or a genuine willingness to learn them); Exposure to Python, LangChain / LlamaIndex, Chunking Strategies is an advantage; Clear written and verbal communication in English, and care with detail in your work; No prior work experience required — this is a fresher-friendly internship; Based in Ghaziabad, Uttar Pradesh or willing to work from there for the internship period.