LLM Engineer's Handbook is finally live!

A framework for building LLM and RAG apps

This is not my regular article.

I want to share my excitement about finally finishing my LLM Engineer's Handbook.

With this book, Maxime and I aim to provide a framework for building LLM and RAG apps!

Consider supporting our work by buying the book from Amazon ←

The journey of writing the book was hardcore.

~ About 4 months ago, Maxime and I started working on the book.

We had to develop code based on constantly shifting AI trends and write 400 pages at an alarming pace.

I probably wouldn't have made it if I hadn't quit my job, and I don't know how Maxime did.

Still, on this tight deadline, we managed to deliver quality and something unique:

"A book that teaches you a framework and mindset on building a production-level LLM and RAG app."

We haven't focused on theory or small projects, as there are other excellent books on the topic.

But we wanted to provide people with an end-to-end experience in building a #GenAI product.

The book will fill in the gaps, showing you how the worlds of DE, SWE, AI, and MLOps work together and highlighting how to solve problems with LLMs and RAG.

The book is split into 3 main sections:

#1. Data:

• Collect and clean custom data

• Feature engineering

• Creating custom datasets for fine-tuning LLMs

• Populating a Qdrant vector DB for RAG

#2. GenAI:

• Fine-tuning and optimizing LLMs using Unsloth

• Advanced RAG

#3. Systems & MLOps:

• Design an LLM system using MLOps & LLMOps best practices

• Deploying the LLM system to AWS SageMaker

• Orchestrate the ML pipelines using ZenML

• Prompt monitoring using Opik from Comet ML

...while everything is connected into a single production-ready project, teaching you how to build your LLM twin.

Also, the diagrams will be in color... and there are many!

Consider supporting our work by buying the book at ↓

I'm super happy to write this book together with Maxime. He is the ninja of fine-tuning LLMs 🔥