LLM Cookbook

Technical guides for building production AI systems.

Guides

Late Chunking vs Contextual Embedding

Michael Ryaboy

The Ultimate Guide to Rerankers

Michael Ryaboy

Choosing an Embedding Model

Michael Ryaboy

The Ultimate Guide to Chunking

Michael Ryaboy

Multi-Stage Reranking

Michael Ryaboy
Join Waitlist

Scaling RAG Systems

Michael Ryaboy
Join Waitlist

About

I'm Michael Ryaboy, a developer advocate at KDB.AI. I build search systems and full-stack AI applications, with a focus on scaling LLMs in production. I've spent the last 5 years implementing LLM-powered applications, and I'm documenting what I've learned here.

These guides cover the technical aspects of working with embedding models, implementing efficient search systems, building reliable AI applications, and more. They're based on practical experience and are intended for engineers working with LLMs in production environments.