Free · No Setup · In Your Browser

Build a RAG app.
Right here. Right now.

Go from zero to a working AI chatbot over your own documents — with interactive demos, real code, and no signup required.

Start Building chevron_right Try the Playground Curriculum
Free forever · No account needed · Open source
LearnRAG Playground

upload Ingest & Embed

"RAG combines retrieval with generation to produce grounded answers..."
"Vector databases store embeddings for fast similarity search..."
"Chunking splits documents into pieces the AI can process..."
check_circle 3 chunks embedded (384 dim)

search Retrieve

How does RAG reduce hallucinations?
#1 0.891
"RAG combines retrieval with generation to produce grounded answers..."
#2 0.743
"Vector databases store embeddings for fast similarity search..."
The Playground — ingest, embed, search, and generate in your browser
The Project

One course. One project. A working RAG chatbot.

Every chapter builds the same thing: a chatbot that answers questions about your own documents — with cited sources and zero hallucinations.

Documents
Ch 1
Chunk
Ch 2
Embed
Ch 3
Store
Ch 4
Search
Ch 5
Answer
Ch 6-8
How It Works
01

Learn the concept

Plain English explanations with zero jargon. Every technical term defined on first use. Real-world analogies before formulas.

02

Try it live

Every concept has an interactive demo. Chunk your own text, explore vector space, compare retrieval strategies — in your browser.

03

Prove you understand

Answer quick checks, earn badges, track your progress locally. No account needed. No grading. Just mastery.

Curriculum

8 chapters. From "what is RAG" to production patterns.

~2.5 hours total. Every chapter has a live interactive demo.

01 Why RAG Exists 12 min Explorer
02 Data Ingestion & Chunking 15 min Explorer
03 Embeddings & Vectors 18 min Builder
04 Vector Databases 14 min Builder
05 Retrieval Strategies 16 min Builder
06 The Prompt Layer 13 min Engineer
07 Evaluation 20 min Engineer
08 Advanced Patterns 25 min Architect
+ 3 hands-on labs with real Python code arrow_forward
The Playground

The fastest way to understand RAG.

Upload text, watch it get chunked, see embeddings form, run similarity search, and generate answers — all in your browser, with no API keys and no setup.

Open the Playground arrow_forward
upload Ingest → search Retrieve → auto_awesome Generate
✓ In-browser embeddings via Transformers.js
✓ No API keys required
✓ See every intermediate step
✓ Failure mode demos included
Open source · Sourced from 13 research papers · Built with Astro + Transformers.js

Ready to build?

No signup. No setup. Just start.

Start Building chevron_right Open Playground
search
construction
bolt
domain
Earn all four badges by building a complete RAG pipeline.