RAG Academy
An open source, AI-guided course that teaches you Retrieval-Augmented Generation by having you actually build it. No vendor lock-in. No outdated tutorials. Just hands-on learning.
About This Project
Most RAG tutorials are either outdated, locked behind paid platforms, or thinly veiled marketing for a specific vendor. RAG Academy takes a different approach. It uses AI as a teaching partner to walk you through every concept, from chunking and embeddings to vector search and generation. Because the teaching is AI-guided, the material stays current as the ecosystem evolves. No deprecated endpoints. No stale code samples. Just the concepts you need to understand, with real code you can run yourself.
Why Build This?
I learned RAG the hard way. Followed tutorials, copy-pasted code, got things running, and still couldn't explain why my retrieval was bad. The understanding only clicked when I started building from scratch and asking "why" at every step. RAG Academy is me trying to make that experience available to everyone. It's my first open source project, and it's free because the resources I learned from were free too.
What's Inside
-
Interactive Lessons
Each lesson focuses on a single concept. Chunking, embeddings, retrieval, generation. You learn one thing at a time and actually understand it before moving on.
-
Hands-on Code
Every concept comes with real code you run yourself. Not just reading. You build, break, and fix things to see how they actually work.
-
AI-Guided Teaching
The LLM isn't just generating answers. It's guiding you through the material, answering your questions, and staying current with the latest tools and best practices.
-
Build Your Own Pipeline
By the end, you'll have a working RAG system you actually understand. Not copy-pasted code, but something you can debug and improve on your own.
Current Status
Early development. Building in public. The project structure and first lessons are taking shape. Follow along on GitHub or check back for updates.