AI Engineering From Scratch
Learn it. Build it. Ship it for others.
What is AI Engineering From Scratch?
AI Engineering From Scratch is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to learn it. build it. ship it for others.
Learn it. Build it. Ship it for others.
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Learn it. Build it. Ship it for others.
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx ai-engineering-from-scratchConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use AI Engineering From Scratch
AI Engineering from Scratch is a structured, hands-on curriculum that teaches machine learning and AI engineering by building concepts from the ground up — from linear algebra through transformers, agents, and production deployment. The course ships as composable agent skills compatible with Claude Code, Cursor, Codex, and other LLM clients, letting AI assistants guide learners phase by phase, run quizzes, and validate understanding. Developers who want to deeply understand the foundations of modern AI systems, not just use them, will find this an end-to-end resource covering math, neural nets, generative AI, multi-agent systems, and MCP server construction.
Prerequisites
- Python 3.10+ installed
- Node.js (for the npx skills installer)
- Claude Code, Cursor, Codex, or another agent that supports skills directories
- API keys for LLM providers (covered in Phase 0, Lesson 04 of the curriculum)
Install all curriculum skills via npx
Use the skills CLI to install the full curriculum into your agent's skills directory. The tool auto-detects the correct directory for Claude Code, Cursor, or other agents.
npx skills add rohitg00/ai-engineering-from-scratchInstall a single skill or phase (optional)
If you want to target a specific phase or skill rather than the full curriculum, use the --skill or --phase flags.
# Install a single skill
npx skills add rohitg00/ai-engineering-from-scratch --skill agent-loop
# Install one phase only
npx skills add rohitg00/ai-engineering-from-scratch --phase 14Install via Python script (offline/custom targets)
Clone the repo and use the Python installer for custom layouts, filtering by tag, or installing into non-standard skill directories.
git clone https://github.com/rohitg00/ai-engineering-from-scratch.git
cd ai-engineering-from-scratch
python3 scripts/install_skills.py ~/.claude/skills/Find your starting phase
Open your AI assistant and run the /find-your-level skill. It administers a ten-question placement quiz and recommends which phase to start from based on your current knowledge.
Work through lessons and check understanding
Complete the lessons in your chosen phase, then run the /check-understanding skill to test what you learned. It gives phase-specific quizzes with feedback and lesson recommendations.
Run lesson code locally
Each lesson includes working Python code. Use lesson_run.py to byte-compile and optionally execute lesson scripts to see the concepts in action.
python3 scripts/lesson_run.py phases/01-math-foundations/01-linear-algebra-intuition/code/vectors.pyAI Engineering From Scratch Examples
Client configuration
This curriculum installs as agent skills rather than a traditional MCP server. Add it to your Claude Code skills directory with the npx command.
{
"mcpServers": {
"ai-engineering-from-scratch": {
"command": "npx",
"args": ["-y", "skills", "add", "rohitg00/ai-engineering-from-scratch"]
}
}
}Prompts to try
Use these skill invocations inside Claude Code or another compatible agent after installing the curriculum.
- "/find-your-level"
- "/check-understanding 3"
- "Explain backpropagation from the Phase 4 materials"
- "Walk me through building a transformer attention layer from scratch"
- "Help me complete the Phase 13 MCP server lesson"Troubleshooting AI Engineering From Scratch
'npx skills' command not found
Ensure Node.js is installed and npx is available (node --version, npx --version). If missing, install Node.js from nodejs.org, then retry the skills add command.
Skills installed but agent doesn't recognize them
Confirm the install target directory matches your agent's skills directory (e.g., ~/.claude/skills/ for Claude Code). Re-run with an explicit path: python3 scripts/install_skills.py ~/.claude/skills/
Lesson Python scripts fail to run
Check the lesson's requirements. Basic math phases only need Python 3.10+, but Phase 3+ introduces PyTorch. Install torch separately (pip install torch) before running those lessons.
Frequently Asked Questions about AI Engineering From Scratch
What is AI Engineering From Scratch?
AI Engineering From Scratch is a Model Context Protocol (MCP) server that learn it. build it. ship it for others. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install AI Engineering From Scratch?
Follow the installation instructions on the AI Engineering From Scratch GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with AI Engineering From Scratch?
AI Engineering From Scratch works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is AI Engineering From Scratch free to use?
Yes, AI Engineering From Scratch is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
AI Engineering From Scratch Alternatives — Similar Developer Tools Servers
Looking for alternatives to AI Engineering From Scratch? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up AI Engineering From Scratch in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use AI Engineering From Scratch?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.