AI Engineering Library
The AI Engineering Library is a series of twelve books for software developers who want to build AI features they can explain and trust. Each book follows one realistic project, such as a pizza shop’s assistant or a banking support assistant, and adds one idea at a time to the same code.
- Local first. The examples run on your own computer through Ollama. There is no account and no paid API. Hosted providers are optional, and no core exercise needs one.
- One real project per book. You start from the simplest working version and move towards a production-aware one.
- Self-contained books. Each book prints the code, setup commands, data and expected output that its exercises need.
Where to start
Section titled “Where to start”If you are new to AI engineering, start with Applied AI Engineering, the entry point of the series. The list below follows volume order, but you do not have to read it in sequence. Each book stands on its own, so you can go straight to the one that covers the topic you need, such as retrieval, agents or security.
- Applied AI Engineering · Pizza Store AI Assistant
- Prompt & Context Engineering · Travel Planning Assistant
- RAG Engineering · Company Knowledge Assistant
- AI Agent Engineering · E-commerce Customer Service Agent
- Model Context Protocol · Personal Productivity Assistant
- AI Harness Engineering · Software Engineering Agent
- Multi-Agent Systems · Logistics and Delivery Coordinator
- AI Agent Workflows · Insurance Claims Processing
- LLMOps · Customer Support AI Platform
- AI Security · Banking Support Assistant
- AI System Design · Enterprise AI Platform
- Multimodal AI · Product Inspection Assistant
All twelve books
Section titled “All twelve books”
Applied AI Engineering
Pizza Store AI Assistant
Follow one project, an assistant for Bella's Pizza, from a first question to a local model to an agent that checks orders and answers from the shop's documents. Every program runs on your own computer through Ollama.
In preparation
Prompt & Context Engineering
Travel Planning Assistant
Build a travel planning assistant for a travel and hospitality business, and learn what to put in a model's request, and why it decides the answer.
In preparation
RAG Engineering
Company Knowledge Assistant
Build a knowledge assistant for a company's documents. It finds the right passages first, then lets the model answer from them.
In preparation
AI Agent Engineering
E-commerce Customer Service Agent
Build a customer service agent for an online shop, one that decides when to use a tool and when to answer directly.
In preparation
Model Context Protocol
Personal Productivity Assistant
Build a personal productivity assistant that connects a model to tools and data through the Model Context Protocol.
In preparation
AI Harness Engineering
Software Engineering Agent
Build a software engineering agent that works inside a controlled environment, with safe file operations, command execution, sandboxing and verification.
In preparation
Multi-Agent Systems
Logistics and Delivery Coordinator
Build a delivery coordinator from several agents that work together, and learn when splitting work across agents helps.
In preparation
AI Agent Workflows
Insurance Claims Processing
Build a claims process that mixes fixed, deterministic steps with agent-assisted steps. Consequential decisions stay under deterministic controls and human review.
In preparation
LLMOps
Customer Support AI Platform
Build a customer support AI platform and learn how to run it in production, with monitoring and evaluation after launch.
In preparation
AI Security
Banking Support Assistant
Build a banking support assistant with fictional data, and learn the security and governance an AI system needs beyond a working prototype.
In preparation
AI System Design
Enterprise AI Platform
Design an enterprise AI platform with a scalable architecture, using realistic architecture without requiring expensive cloud infrastructure.
In preparation
Multimodal AI
Product Inspection Assistant
Build a product inspection assistant that works with images as well as text, using local models whose abilities have been verified.
In preparation
Book text © 2026 Ravi Pal. All rights reserved. Code examples are licensed under the MIT License.