ai-engineering-from-scratch
View on GitHubLearn it. Build it. Ship it for others.
Open-source 523-lesson, 20-phase AI engineering curriculum that builds math, transformers, LLM apps, agents, MCP and Agent Skills from scratch in Python, TypeScript, Rust and Julia. Ships installable Agent Skills (npx skills add) so a coding agent can tutor you lesson by lesson.
Use Cases
Learn AI engineering fundamentals from scratchBuild LLM applications and prompt pipelinesBuild AI agent loops and multi-agent swarmsLearn and implement Model Context Protocol (MCP)Author and ship Agent Skills for coding agentsImplement transformers, tokenizers and backprop by handPrepare for Claude certification examsStudy math foundations for MLPractice evaluation and reliability engineering for LLMsUse coding agents on real repositoriesFollow structured learning paths in Python, TypeScript, Rust, Julia
Built With
- Language
- Python
- Frameworks
- PyTorch · Transformers · NumPy · scikit-learn · pandas · Jupyter · Matplotlib · Hugging Face Datasets · Tokenizers · Accelerate · OpenAI SDK · Anthropic SDK · librosa · tiktoken
Tags
curriculum · education · from-scratch · ai-engineering · agents · mcp · agent-skills · llm-engineering · transformers · deep-learning · machine-learning · reinforcement-learning · prompt-engineering · computer-vision · nlp · multi-language