Building AI Agents That Ship

Tools, memory, planning and guardrails

4.9 7,340 students 3h 30m
Advanced
Free course
An
Taught by Andrej Karpathy
Lessons play via the official YouTube embed. All views credit the creator.
Visit channel

Create a free account to start learning

Every course is free — you just need an account so we can save your progress, bookmarks and certificates.

Anatomy of an agent loop

Video by Andrej Karpathy · Watch on YouTube

AI lesson summary

Plan, act, observe, repeat: the core loop and where agents typically break.

Key takeaways

  • Tools turn a chat model into an actor
  • State must be explicit and inspectable
  • Stop conditions prevent runaway cost

Practical exercise

Diagram an agent that books a meeting, including failure paths.

About this course

Design agent loops that actually finish tasks: tool calling, state, retries, evaluation and cost control.

agents
tools
automation

Knowledge check

Generate a free AI quiz to test what you just learned.

Notes & study tools