Curriculum
In the 2026 AI Dev Tools Zoomcamp, you complete five modules plus a final project. You follow one application from specification through development and deployment. You then add operations and agent extensions.
For the canonical curriculum (videos, code, exact homework), see the GitHub repository.
Modules
Module 1: AI-Native Developer Workflow
- Compare AI developer tools, turn an idea into a specification and backlog, give agents durable project context, and verify their work independently.
Module 2: Build and Ship an AI-Assisted Full-Stack App
- Write a product specification, build a frontend, define the API with OpenAPI, implement the backend, add persistence, and test the main behavior.
Module 3 - Test, Containerize, and Deploy
- Add integration tests and move from SQLite to Postgres. Package the application with Docker, run checks in CI, and deploy from the main branch.
Module 4: DevOps and Observability for AI-Built Apps
- Connect metrics, logs, and traces, alert on user impact, investigate incidents with bounded agent access, and run recurring security audits.
Module 5: Coding Agent Capabilities
- Extend coding agents with project instructions, MCP, reusable skills, hooks, specialized subagents, plugins, and custom agents. You also document permissions and guardrails.
- Build and deploy a full-stack application of your own. Document how you directed and reviewed AI tools, then add the operations, security, and agent-extension work covered in the later modules.
Homework and project
We score homework during the live cohort and show it on the leaderboard, but it isn’t required for the certificate. To earn the certificate, submit a passing final project and complete the required peer reviews during the live cohort. See Project for the course requirements and the course platform for exact deadlines.