AutoGPT is an open-source platform for building, running, and scheduling AI agents that carry out multi-step workflows from plain-English descriptions or a visual builder.
187.7K45.9KPythonOtherUpdated yesterday
rohitg00/ai-engineering-from-scratch
A free, MIT-licensed course of 542 lessons that builds AI engineering concepts from scratch in Python, TypeScript, Rust and Julia.
- Stars
- 71,373
- Forks
- 12,556
- Open issues
- 75
- Last commit
- yesterday
What is it?
AI Engineering from Scratch is a structured course covering the full stack of modern AI work, from math foundations and classical machine learning through deep learning, computer vision, NLP, LLM engineering, and agent systems. The lessons are organized into 20 phases, and each one pairs a written explanation with code that the learner builds and runs locally.
The course targets developers who want to understand how models and agent runtimes actually work rather than only calling APIs. Lessons ask learners to implement internals such as backpropagation, retrieval pipelines and tool-calling layers, then keep the outputs as evidence. Lessons can be followed on the website, through a coding agent installed with npx skills, or by cloning the repository and running the Python scripts directly.
The repository also includes three staged projects with local graders: a retrieval evaluation lab, an agent trace debugger, and a tool call firewall. Starter code fails until learners implement each stage, and reference solutions are provided for comparison.
Key features
- 01542 lessons across 20 phases, from setup and math to agent engineering
- 02Lessons pair written explanations with code to build and run locally
- 03Supports Python, TypeScript, Rust and Julia across different lessons
- 04Includes an interactive gradient descent visualization and optimizer comparison code
- 05Three staged projects with local graders and reference solutions
- 06Optional AI tutor skills for placement quizzes, per-phase quizzes and personalized study plans
Use cases
- Developers who want to understand neural network internals by implementing them
- Engineers moving into LLM application or agent development who want a structured path
- Teams building retrieval evaluation or agent debugging skills with hands-on projects
- Learners who prefer working through code and saving verifiable outputs
Quick start
- 1
Clone the repository
git clone https://github.com/rohitg00/ai-engineering-from-scratch.git - 2
Install the AI tutor skill for your coding agent
npx skills add rohitg00/ai-engineering-from-scratch - 3
Check that Python 3 is available for the local labs
python3 --version
Things to know
- Many lessons are meant to be read on the website, so local running is not required to learn from it.
- The project challenges need Python 3.10+ or Node.js 22.18+ depending on the project.
- Agent Skills labs need a supported coding agent host and a writable skill scope.
Topics
FAQ
- What is ai-engineering-from-scratch?
- A free, MIT-licensed course of 542 lessons that builds AI engineering concepts from scratch in Python, TypeScript, Rust and Julia. AI Engineering from Scratch is a structured course covering the full stack of modern AI work, from math foundations and classical machine learning through deep learning, computer vision, NLP, LLM engineering, and agent systems. The lessons are organized into 20 phases, and each one pairs a written explanation with code that the learner builds and runs locally.
- Is ai-engineering-from-scratch free to use? What license does it have?
- ai-engineering-from-scratch is open source under the MIT license. Check the license terms before using it in commercial products.
- Is ai-engineering-from-scratch still maintained?
- Actively maintained. The last commit to ai-engineering-from-scratch was on Oct 19, 2026.
- What can I use ai-engineering-from-scratch for?
- Developers who want to understand neural network internals by implementing them Engineers moving into LLM application or agent development who want a structured path Teams building retrieval evaluation or agent debugging skills with hands-on projects Learners who prefer working through code and saving verifiable outputs
- How do I install ai-engineering-from-scratch?
- Clone the repository: git clone https://github.com/rohitg00/ai-engineering-from-scratch.git Install the AI tutor skill for your coding agent: npx skills add rohitg00/ai-engineering-from-scratch Check that Python 3 is available for the local labs: python3 --version
- Where can I get ai-engineering-from-scratch?
- The source code and installation instructions are on GitHub: https://github.com/rohitg00/ai-engineering-from-scratch
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A Python framework for building LLM-powered applications and agents, with standard interfaces for models, embeddings, vector stores and third-party integrations.
148.1K24.9KPythonMITUpdated 4 hours ago
Open-source Python framework for orchestrating role-based autonomous AI agents (Crews) and event-driven workflows (Flows).
59.9K8.8KPythonMITUpdated 5 hours ago
A low-level Python framework for building stateful, long-running agents with durable execution, human-in-the-loop and memory.
43.6K7.4KPythonMITUpdated 5 hours ago
A Python framework and runtime for building, serving, and managing AI agent platforms, with a web UI for administration.
42.8K6.2KPythonApache-2.0Updated yesterday
Python framework for building autonomous AI agents that research, plan, code and run tasks, with memory, RAG and 100+ LLM providers.
9.3K1.5KPythonMITUpdated 3 hours ago