# rohitg00/ai-engineering-from-scratch

Learn it. Build it. Ship it for others.

Repository: https://github.com/rohitg00/ai-engineering-from-scratch
Canonical: https://ross.abutalabs.com/products/ai-engineering-from-scratch
Homepage: https://aiengineeringfromscratch.com
Language: Python
License: MIT
License Family: permissive
Topics: agents, ai, ai-agents, ai-engineering, computer-vision, course, deep-learning, from-scratch, generative-ai, llm, machine-learning, mcp, nlp, python, reinforcement-learning, rust, swarm-intelligence, transformers, tutorial, typescript
Last push: 2026-08-23T20:29:51+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 97, longevity 12
- inputs: {"age_days": 168, "days_push": 10, "days_rel": 23, "gap_med": 15, "n_releases_24m": 2}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 49577, forks 8628 (observed 2026-08-28T04:12:11.612053+00:00)

## What it is
A free, MIT-licensed open-source curriculum of 511 lessons across 20 phases (~329 hours) that teaches AI engineering by building every core algorithm from raw math before using frameworks. Lessons span linear algebra to autonomous agents in Python, TypeScript, Rust, and Julia, each producing a reusable artifact like a prompt, skill, agent, or MCP server.

## Use cases
- learn AI engineering from scratch
- understand how transformers and attention work internally
- build agents and MCP servers by hand
- study backpropagation and tokenizers from raw math
- find a structured alternative to scattered AI tutorials
- prepare for a professional AI engineering role
- learn deep learning without frameworks first

## When to choose
- you want deep foundational understanding of AI algorithms, not just framework usage
- you prefer self-paced, free, open-source learning material
- you want lessons that produce reusable artifacts like prompts, skills, and MCP servers
- you want curriculum usable with AI coding agents via SKILL.md

## When to avoid
- you need a quick framework-specific tutorial like 'fine-tune with PyTorch in an hour'
- you want instructor-led or accredited courses with certificates
- you need production-ready software rather than educational material

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, nlp, agent-framework, mcp, prompt-engineering, computer-vision, llm-training
- domain: artificial-intelligence, machine-learning, deep-learning, large-language-models, education, tutorials
- platform: python, rust, cross-platform, cli
- tags: curriculum, from-scratch, generative-ai, transformers, agents, mcp-servers, self-paced-learning, open-courseware, ai-agents

## Member repositories
- rohitg00/ai-engineering-from-scratch (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:11.612053+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:20:11.557143+00:00, confidence not recorded.
  - readme: https://github.com/rohitg00/ai-engineering-from-scratch (fetched 2026-08-28T04:12:11.612053+00:00, sha 27df0351d445)
  - homepage: https://aiengineeringfromscratch.com (fetched 2026-08-28T18:13:09.024015+00:00, sha 9fe4dce9f0b2)
  - site_page: https://aiengineeringfromscratch.com/about.html (fetched 2026-08-28T18:13:09.033107+00:00, sha 41935946900c)
- Data as of 2026-08-30T08:39:29.467469+00:00.