AI learning paths

Practical routes through AI learning.

Start from the work you need to do. Each path gives you the concepts to learn, useful resources, and a rough sequence.

Path 3

Automation and agents

Founders, operations teams, developers

Learn first

  1. Trigger-based workflows
  2. Tool calling
  3. Agent failure modes
  4. Human review points

Use these names

  • Brian Casel
  • Josh Pigford
  • AI Engineer
Path 4

AI product work

PMs, designers, founders

Learn first

  1. AI UX patterns
  2. Use-case selection
  3. Prototyping
  4. Measuring quality

Use these names

  • Peter Yang
  • Lenny Rachitsky
  • Shreya Shankar
Path 5

Coding with AI

Software developers

Learn first

  1. Codebase navigation
  2. Test generation
  3. Refactoring
  4. Reviewing AI output

Use these names

  • Simon Willison
  • Addy Osmani
  • AI Engineer
Path 6

RAG and knowledge systems

Developers building AI apps

Learn first

  1. Chunking and retrieval
  2. Structured extraction
  3. Reranking
  4. Grounded answers

Use these names

  • Jason Liu
  • Hamel Husain
  • Eugene Yan
Path 7

Evals and reliability

AI product teams

Learn first

  1. Test sets
  2. Human review
  3. Regression checks
  4. Quality metrics

Use these names

  • Hamel Husain
  • Shreya Shankar
  • Chip Huyen
Path 9

Production AI engineering

Engineering teams

Learn first

  1. Observability
  2. Cost and latency
  3. Deployment
  4. Data pipelines

Use these names

  • Full Stack Deep Learning
  • Chip Huyen
  • Eugene Yan

Open next

Path 10

Strategy and adoption

Leaders, managers, founders

Learn first

  1. Team habits
  2. Org workflows
  3. Policy
  4. Change management

Use these names

  • Ethan Mollick
  • Lenny Rachitsky
  • Latent Space