ai-engineering-from-scratch
Learn AI Engineering! Learn it. Build it. Ship it for others.
What is ai-engineering-from-scratch?
Learn AI Engineering! Learn it. Build it. Ship it for others.
Overview of ai-engineering-from-scratch
What it is
ai-engineering-from-scratch is a free, open-source curriculum for learning AI engineering. It covers model internals, retrieval pipelines, and agent runtimes. The project includes 542 lessons across 20 phases, using Python, TypeScript, Rust, and Julia. It is licensed under MIT and hosted on GitHub at https://github.com/rohitg00/ai-engineering-from-scratch, with a companion website at https://aiengineeringfromscratch.com.
Key features
- 542 lessons organized into 20 phases, covering model foundations, LLM systems, and agents and delivery.
- Three projects with staged starters, reference implementations, and local graders: Retrieval Evaluation Lab (Python), Agent Trace Debugger (TypeScript), and Tool Call Firewall (Rust).
- Learning paths for model foundations, LLM systems, and agents and delivery, with a placement quiz to map existing knowledge.
- Can be learned on the website, with an AI tutor via a coding agent, or by running local code.
- Supports multiple languages for the README, including English, Simplified Chinese, Traditional Chinese (Taiwan), Japanese, Korean, and many others.
- Includes a tutor setup using
npx skills add rohitg00/ai-engineering-from-scratchfor compatible coding agents.
Use cases
Learning AI engineering from scratch, including model internals, retrieval pipelines, and agent runtimes. Suitable for self-study, classroom use, or as a reference for building and testing AI systems. The projects can be used to practice ranking metrics, trace debugging, and tool call security.
Getting started
To learn on the website, open any completed lesson at aiengineeringfromscratch.com. To use the AI tutor, ensure Node.js, npx, and a compatible coding agent are installed, then run:bashnpx skills add rohitg00/ai-engineering-from-scratchFor local code, clone the repository and follow the instructions in the README. See the repository README for detailed setup and project commands.
License and pricing
This project is open source under the MIT license. It is free to use.
Based on the project's public README and repository metadata.