Build production-grade RAG systems and AI agents with hybrid search, evals, LangGraph, MCP, guardrails and observability, with AI-reviewed code on GitHub
An advanced 8-week, project-based program for developers who already know Python and have called an LLM API. You go deep on the two architectures behind most AI products in 2026: retrieval-augmented generation and AI agents. You master retrieval with embeddings, chunking, pgvector and Qdrant, hybrid BM25 search and rerankers, then advanced RAG with query rewriting, HyDE, contextual retrieval, GraphRAG, multimodal PDFs and verified citations. You measure everything with golden datasets, recall@k, Ragas, calibrated LLM judges and tracing. You then build agents from scratch with tool calling, ship them with LangGraph, agent SDKs and MCP, and harden them with guardrails, prompt-injection defences, sandboxes, agent evals and cost controls. Every week ends with a project on GitHub reviewed by AI, and you finish with a capstone and interview prep.
Category: Generative AI
Level: Advanced
Length: 8 weeks, online and self-paced
Price: ₹5,999
Rating: 8.5/10
What you will learn
Use Git and GitHub with branches and pull requests, the way professional AI teams work
Choose embedding models and chunking strategies by measuring them on your own data
Build ingestion pipelines with rich metadata and incremental re-indexing
Store and query vectors in pgvector, Qdrant and Chroma with metadata filters
Combine BM25 and vector search with rank fusion and add cross-encoder rerankers
Apply query rewriting, multi-query, HyDE, routing and contextual retrieval
Handle PDFs with tables and images and explore GraphRAG for multi-hop questions
Produce grounded answers with verified citations and honest refusals
Evaluate RAG with golden datasets, recall@k, MRR, Ragas, DeepEval and calibrated LLM judges
Trace and tune pipelines with Langfuse or LangSmith for quality, cost and latency
Build agent loops from scratch with Claude and OpenAI tool calling, planning and memory
Choose between workflows and agents using Anthropic's building-effective-agents patterns
Build agents with LangGraph, the Claude Agent SDK, the OpenAI Agents SDK and CrewAI
Build and consume Model Context Protocol servers with human-in-the-loop approval
Secure agents with guardrails, prompt-injection defences, sandboxes and agent evals
Deploy agents with FastAPI, background queues, observability and cost controls
Skills
Version control with Git and GitHub
Embeddings and chunking strategies
Vector databases: pgvector, Qdrant and Chroma
Hybrid search and reranking
Advanced RAG techniques
Multimodal RAG and GraphRAG
RAG evaluation with Ragas and DeepEval
LLM-as-judge and golden datasets
LLM observability with Langfuse and LangSmith
Tool calling with Claude and OpenAI APIs
Agent design patterns, planning and memory
LangGraph and agent SDKs
Model Context Protocol (MCP)
Agent guardrails and prompt-injection defence
Deploying agents with FastAPI and queues
Frequently asked questions
Do I need prior experience?
This is an advanced course. You should be comfortable with Python and have built at least a basic app that calls an LLM API.
How long is the course and how much time do I need each week?
The course runs for 2 months, with 8 weekly modules. Most learners spend 8–10 hours a week. It is self-paced, so you can go faster or slower.
How are my assignments reviewed?
You push your code to GitHub and AI reviewers check it for bugs, quality and security within minutes, with clear fixes.
Will I get a certificate?
Yes. Finish the weekly assignments and the final project to download a verified certificate with a unique ID that employers can check.
Can I ask questions if I get stuck?
Yes. Use Ask your mentor inside any week to send a question to your mentor.
What roles does this course prepare me for?
RAG & AI Agents Engineering prepares you for roles such as AI Engineer, LLM Application Engineer, AI Agent Developer and Applied ML Engineer (GenAI). The projects you push to GitHub become a portfolio you can show in interviews.