DataTalksClub/ai-dev-tools-zoomcamp

A free, hands-on course on using AI developer tools to build, test, deploy, extend, and audit software without losing engineering discipline. The 2026 cohort starts August 31.

What is it?

A free online course from DataTalks.Club that teaches developers how to use AI coding tools (like coding assistants and agents) across the whole software lifecycle — from planning, to building, testing, deploying, and even investigating bugs. It is explicitly not about prompt tricks, model training, or RAG systems. It is about day-to-day programming with AI involved, but with guardrails.

Core workflow taught

The course is organized around one loop: give AI tools the right context, use them for the right job, review what they produce, test the result, and ship with guardrails.

Who it's for

You should already write basic code (Python, JavaScript, TypeScript, or similar), be comfortable with the command line, and know Git/GitHub basics. It's relevant to software engineers, ML engineers, data scientists, data engineers, analysts who write code, and technical students who want practical project experience. It is not for beginners who have never programmed, or for people who mainly want LangChain, vector databases, or RAG.

What you'll build and learn (5 modules)

  1. AI-Native Developer Workflow — turn a vague idea into a spec and backlog, give coding agents context (AGENTS.md), and use PM/engineer/QA roles to implement and verify.
  2. Build and Ship an AI-Assisted Full-Stack App — build a frontend prototype, define an OpenAPI contract, implement a FastAPI backend, add auth, real-time collaboration, SQLite storage, and tests.
  3. Test, Containerize, and Deploy — containerize the app, move to Postgres with Docker Compose, add integration/end-to-end tests (including Playwright), deploy to AWS, and set up GitHub Actions CI/CD.
  4. DevOps and Observability for AI-Built Apps — separate dev/prod environments, instrument with OpenTelemetry into Prometheus/Loki/Tempo/Grafana, set up alerts, and give a coding agent a read-only incident-responder role.
  5. Coding Agent Building Blocks — create reusable skills (SKILL.md), define focused subagents, and orchestrate parallel work using Git worktrees. (No graded homework here — assessed via a deliverable instead.)

How the course works

  • Live cohort (starts August 31, 2026): free, pre-recorded lectures, but with shared deadlines, graded homework, a leaderboard, peer review, community support, and certificate eligibility.
  • Self-paced: free, use materials and videos anytime, homework available but not scored, no leaderboard/peer review/certificate.

Both options are free. Lectures are pre-recorded either way.

Certificate

Certificate eligibility requires participating in a live cohort. You must complete the final project (build, test, containerize, and deploy your own end-to-end app) and complete required peer reviews. Homework helps you stay on track but does not determine the certificate.

Final project

You pick a problem, then: build a frontend, build a backend, define an OpenAPI contract, add persistent storage, test, containerize, deploy it publicly, and document how you used AI tools/prompts/agents/MCP/automation.

Instructors

Alexey Grigorev and Bhavani Ravi.

Community

dataTalks.Club Slack (#course-ai-dev-tools-zoomcamp), a Telegram channel for announcements, and a learning-in-public hashtag (#aidevtools).

One caveat

The 2026 materials are still being finalized. The repository shows the direction for the next cohort, but some videos, homework, deadlines, and project requirements may change before the August 31, 2026 start date.

Related

  • Project
    cranebeaker/coop-anything

    Co‑op Anything is an MIT‑licensed framework that uses LLM‑driven coding agents to turn any PC single‑player game into a multiplayer‑ready co‑op mod. It provides playbooks, agent skills, and an MCP server so agents can research, reverse‑engineer, test with bots, and package a one‑click Steam‑relay install. Existing co‑op mods are revived or new ones built from scratch for a wide range of engines, with several games (Deus Ex, Dungeon Keeper, S.T.A.L.K.E.R., Fallout NV, Ocarina of Time) already supported.

  • Project
    spec-kitty/spec-kitty

    Spec Kitty is an open-source command-line tool that turns product intent into a structured, spec-driven development workflow for AI coding agents. It keeps specs, plans, work packages, acceptance criteria, and merge decisions in your repository, and isolates agent work in git worktrees so multiple agents can work in parallel without branch conflicts.

  • Project
    fynnfluegge/agtx

    A terminal-native agentic development environment that uses a blackboard system to coordinate multiple AI coding agents across a shared Kanban board.

  • Project
    gi-dellav/zerostack

    zerostack is a lightweight coding agent written in Rust that runs in the terminal, supporting multiple LLM providers. It offers granular permission controls, session management, subagents, and optional persistent memory, while using far less RAM and CPU than JavaScript-based agents.