ai-engineering-from-scratch
Learn it. Build it. Ship it for others.
ai-engineering-from-scratch is a comprehensive learning curriculum that teaches developers to build AI systems end-to-end from mathematical foundations. Structured as 503 lessons across 20 phases, it covers everything from linear algebra through autonomous multi-agent systems, with hands-on implementations in Python, TypeScript, Rust, and Julia. Each lesson produces a reusable artifact—prompts, skills, agents, or MCP servers—turning theoretical knowledge into practical tools deployable in production workflows.
Key Features
Use Cases
- 01Learning AI engineering from first principles without prior machine learning background
- 02Building production-ready AI agents and MCP servers with deep understanding of underlying algorithms
- 03Preparing for Claude Associate, Developer, Architect Foundations, and Architect Professional certifications
- 04Creating a portfolio of 503 reusable AI tools, prompts, and skills for daily development workflow
- 05Training coding agents to become interactive AI tutors for personalized learning
- 06Transitioning from AI API consumer to AI system builder who understands backpropagation, attention mechanisms, and agent loops
Related Agents
View moreunsloth
Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.
codegraph
Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local
claude-code-best-practice
from vibe coding to agentic engineering - practice makes claude perfect
system_prompts_leaks
Extracted system prompts from Anthropic - Claude Fable 5, Opus 5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-5.6-Sol, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.
ai-engineering-from-scratch — FAQ
What is ai-engineering-from-scratch?+
It is an open-source AI engineering curriculum that teaches you to build AI systems from mathematical foundations through production deployment. The program delivers 503 lessons via interactive AI agent skills, producing reusable artifacts like prompts, agents, and MCP servers.
How do I install ai-engineering-from-scratch in my coding agent?+
Run npx skills add rohitg00/ai-engineering-from-scratch in your terminal, then use /start-learning inside your agent to begin. The installer supports Claude Code, Cursor, Codex, OpenClaw, Hermes, and any agent that reads SKILL.md directories.
Which AI clients work with ai-engineering-from-scratch?+
The learning skills work with Claude Code, Cursor, Codex, OpenClaw, Hermes, and any coding agent that supports the SKILL.md format. You can also read lessons at aiengineeringfromscratch.com or clone the repository directly.
What are the prerequisites for using this curriculum?+
You need basic coding ability in any language (Python helps) and a desire to understand how AI actually works, not just call APIs. No prior machine learning knowledge is required; the course starts with linear algebra intuition.
Is ai-engineering-from-scratch free to use?+
Yes, the entire curriculum is free and open source under the MIT license. All 503 lessons, code implementations, skills, and book volumes are available at no cost.
Do I need API keys to use ai-engineering-from-scratch?+
No API keys are required for the core curriculum lessons, which teach you to build algorithms from scratch. However, later lessons involving LLM integration and agent building will use whatever AI client you've already configured.
How do I install ai-engineering-from-scratch?+
Open the source repository on GitHub and follow its README. ai-engineering-from-scratch is a agent — MCP Agents Market links you directly to the official repo.
Is ai-engineering-from-scratch free?+
ai-engineering-from-scratch is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.