RAG Workshop: Build AI Apps That Know Your Data
Science And Technology

RAG Workshop: Build AI Apps That Know Your Data

A hands-on workshop on Retrieval-Augmented Generation. Learn to connect Large Language Models to your own documents and build smart, accurate AI assistants from scratch.
Course ModeOnline
Duration2 Weeks
Starts18 Jan 2027
Ends30 Jan 2027
Course Fee
₹1,000/-
Enroll Now
About The Course

Overview

Large Language Models like GPT, Claude and Gemini are powerful, but they don't know about your company's data, your notes or anything after their training cutoff, and they sometimes make things up. Retrieval-Augmented Generation (RAG) solves this by fetching the right information from your own data and giving it to the model before it answers. In this workshop you will go from the basics of embeddings and vector databases to building and deploying a complete RAG application. Every concept is followed by live coding, so you leave with a working project, not just theory. It is designed for students, developers and AI enthusiasts with basic Python knowledge. No prior machine learning experience is required.

Learning Goals

Course Outcomes

By the end of this workshop, you will be able to: • Explain how RAG works and why it makes LLM answers more accurate and reliable • Convert raw documents into embeddings and store them in a vector database • Apply the right chunking and retrieval strategies for different types of data • Build a complete RAG pipeline using Python, LangChain / LlamaIndex and an LLM API • Improve results with hybrid search, re-ranking and query transformation • Evaluate a RAG system for relevance, accuracy and hallucinations • Deploy a working document Q&A chatbot you can add to your portfolio

Curriculum

Course / Modules

Course Modules Module 1: Foundations of LLMs and Why RAG Exists • How large language models generate answers • Limitations: hallucinations, knowledge cutoff, no access to private data • RAG vs fine-tuning vs prompt engineering: when to use which Module 2: Embeddings and Vector Search • What embeddings are and how semantic similarity works • Choosing an embedding model • Vector databases: Pinecone, ChromaDB, FAISS, pgvectorModule 3: Data Ingestion and Chunking • Loading data from PDFs, websites, docs and databases • Chunking strategies: fixed-size, recursive, semantic • Adding metadata for better filtering and retrieval Module 4: Building Your First RAG Pipeline • Retriever + prompt + LLM: the core RAG flow • Hands-on with LangChain / LlamaIndex • Writing prompts that ground answers in retrieved context Module 5: Advanced Retrieval Techniques • Hybrid search (keyword + vector) and re-ranking • Query rewriting, multi-query and HyDE • Conversational RAG with chat memory Module 6: Evaluation, Optimization and Deployment • Measuring retrieval quality and answer faithfulness (RAGAS) • Reducing latency and cost • Deploying a RAG app with a simple API and UI Capstone Project • Build a "Chat with your Documents" assistant end to end • Upload files, ask questions and get answers with source citations

Learn From Experts

Our Trainers

Tanishq Vyas
Tanishq Vyas
Trainer at Sheriyan Coding School