Introduction

The AI Engineer Handbook is a practical, hands-on guide to AI Engineering. It aims to provide programmers a systematic introduction into building applications which use Large Language Models (LLMs) such as ChatGPT, GPT-4, Anthropic Claude and others.

When providing code examples, care is taken to avoid the use of frameworks and instead providing code which demonstrates the high-level flow so that you can implement it yourself (or use a framework if you so choose).

Pre-requisites

This handbook is designed for programmers in mind, in particular programmers who can read/write Python. No prior machine learning or AI knowledge is assumed.

What is AI Engineering?

Accoring to @karpathy of OpenAI:

Key takeaways:

• AI engineers use LLMs rather than build them

• They operate at a higher level of abstraction than ML engineers or LLM engineers

• AI engineers don't need to know how to build an LLM or a ML model, though it does help to know how they work

• There are some distinct skills which AI engineers do need to know such as:

  • Prompt engineering

  • Working with data

  • LLM infra

  • Evaluating LLMs

Swyx has more on this: The Rise of the AI Engineer

Orientation

This handbook follows the Diátaxis documentation framework.

If you are here to learn, go to tutorials.

HOWTOs on the other hand, are task-oriented and exist to help you problem-solve and accomplish a task.

The theory section is useful when you are looking to dive deeper into how things work.

The reference section provides a glossary.