Become an AI engineer

# Retrieval and tools

Retrieval and tools is a unit of 9 lessons in Rabbithole's Become an AI engineer course, at levels 3 to 4.

## Lessons

1. BM25: search that matches the exact words · Level 3
2. Chunking: cutting text into pieces that retrieve well · Level 3
3. Hybrid Search with MongoDB $rankFusion · Level 3
4. Knowledge Graphs for RAG · Level 3
5. Retrieval-Augmented Generation · Level 3
6. Vector Search · Level 3
7. Reranking: re-scoring the short list · Level 3
8. Contextual Retrieval: Giving Each Chunk Its Context · Level 4
9. ReAct: thinking and acting in one loop · Level 4

## Sources

These lessons draw on 18 sources. The most used:

- [Claude Cookbooks](https://github.com/anthropics/claude-cookbooks)
- [Supabase Documentation](https://supabase.com/docs)
- [openai/openai-cookbook](https://github.com/openai/openai-cookbook)
- [Improving language models by retrieving from trillions of tokens](https://arxiv.org/abs/2112.04426)
- [Anthropic Engineering blog](https://www.anthropic.com/engineering)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909)
- [Hugging Face Agents Course](https://github.com/huggingface/agents-course)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629)

[All units in Become an AI engineer](https://rabbithole.pro/courses/ai-engineer)

Source page: https://rabbithole.pro/courses/ai-engineer/retrieval-and-tools