# ai-engineering-from-scratch

> ai-engineering-from-scratch is an open-source AI tool for ai models and llm tooling. Free open-source course that builds AI systems from raw math upward. It is available on Web and CLI.

- **Kind:** Tool
- **Pricing:** Open source
- **Platforms:** Web, CLI
- **Website:** https://aiengineeringfromscratch.com
- **Repository:** https://github.com/rohitg00/ai-engineering-from-scratch
- **GitHub stars:** 79,174
- **License:** MIT
- **Categories:** AI models and LLM tooling, research
- **Listing updated:** 2026-11-14

## About

The course site lists 542 lessons across math, models, tokenizers, agents, evaluation and production, each repeating a loop of problem, derivation, code, test and kept artifact. Code is given in Python, TypeScript, Rust and Julia, and algorithms such as backprop, tokenizers, attention and the agent loop are written by hand before any framework is used.

A command installs the lessons as a skill so an agent can act as tutor in the terminal, with a placement quiz and a personalised path. Learning paths include AI engineering, agent harness work, stateless MCP systems, portable skills and Claude or MCPA certification routes.

It is built to run on your own machine and is MIT licensed.

## Features

- **542 lessons:** Covers math, models, tokenizers, agents, evaluation and production.
- **Four languages:** Lessons are implemented in Python, TypeScript, Rust and Julia.
- **Agent as tutor:** A skill command turns your agent into a terminal tutor with a placement quiz.
- **Learning paths:** Separate routes for AI engineering, MCP systems and skills.
- **Kept artifacts:** Each lesson ends with an artifact you keep.

## Good for

- Learning how attention and backprop work by implementing them
- Studying how to build and secure MCP systems
- Following a guided path toward an AI engineering role
- Practising skill authoring in a real agent host

## FAQ

### Is it free?

The site describes it as free and open source, under the MIT licence.

### How do I start?

Run npx skills add rohitg00/ai-engineering-from-scratch and use start-learning.

### Which languages are used?

Python, TypeScript, Rust and Julia.

### Is ai-engineering-from-scratch free?

Yes. ai-engineering-from-scratch is open source, so you can use and self-host it for free.

### What are the best alternatives to ai-engineering-from-scratch?

The closest alternatives to ai-engineering-from-scratch are GPT-3 Playground (OpenAI), ChatGPT (OpenAI), to teach, Heyday and ECC.

## Alternatives

- [GPT-3 Playground (OpenAI)](https://ai-powered.com/tools/gpt-3_playground_(openai)): GPT-3 Playground (OpenAI) is a free AI tool for copywriting.
- [ChatGPT (OpenAI)](https://ai-powered.com/tools/chatgpt_(openai)): ChatGPT (OpenAI) is a free AI tool for chat.
- [to teach](https://ai-powered.com/tools/to_teach): to teach is a freemium AI tool for productivity.
- [Heyday](https://ai-powered.com/tools/heyday): Heyday is an AI tool for productivity.
- [ECC](https://ai-powered.com/tools/ecc): ECC is a freemium AI tool for ai models and llm tooling.
- [Teach Anything](https://ai-powered.com/tools/teach_anything): Teach Anything is a free AI tool for research.
- [Perplexity AI](https://ai-powered.com/tools/perplexity_ai): Perplexity AI is a free AI tool for chat.
- [ponytail](https://ai-powered.com/tools/ponytail): ponytail is an open-source AI tool for ai models and llm tooling.

---
Source: https://ai-powered.com/tools/ai-engineering-from-scratch
