AI Dev Tools Zoomcamp

Louis Amaudruz built a team of AI agents in Slack, live, during our latest workshop. They worked together on GitHub issues for the Switch repository.

The recording is now up.

Louis started with an empty Slack workspace and walked through each agent:

- A Claude Code coder agent that fixes an issue and opens a pull request
- A Codex reviewer agent that reviews the fix in the same channel
- A lead agent that runs the code-review loop, with a maximum of two cycles
- An issue manager that creates one channel per issue and starts the work in each of them

Louis also showed how the Switch team uses the same pattern for its own work, with rooms for bug reports, feature work, local deployments, releases, and staging deployments.

Watch the recording to follow the full build:
https://www.youtube.com/watch?v=vl00W8eEtZg

We start Module 4 on DevOps and Observability for AI-Built Apps.

Once your application is deployed, you need to know what it is doing. With observability, you can see where the app failed, investigate with the right context, and decide how to respond.

We cover:

- What to observe in a running application
- Which failures should trigger alerts
- What context an investigation needs
- Where coding agents can help with operations safely
- Which production permissions stay with humans

Materials:
https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/tree/main/04-devops

Recording:
https://www.youtube.com/watch?v=YkxLo_FRoQw

Companion article:
https://aishippingblog.com/p/devops-and-observability-for-an-ai

Homework 4 is currently a draft and the questions will likely change. We will announce once it’s ready.

Homework:
https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/blob/main/cohorts/2026/04-devops/homework.md

Submit:
https://courses.datatalks.club/ai-dev-tools-2026/homework/hw4

Deadline: September 29, 2026 at 01:00 CET (September 28 at 23:00 UTC).

Next week, the final module covers reusable coding-agent capabilities: skills and specialized subagents.

Questions go in the course Slack channel:
https://app.slack.com/client/T01ATQK62F8/C09HWT76L95

New free model:

https://x.com/OpenRouter/status/2100235351575191751

We're starting a stream about organizing a team of agents and creating a startup

https://www.youtube.com/watch?v=vl00W8eEtZg

Join us here or watch later in recording

We start Module 3

Getting an application to work once is different from making it reproducible and ready for other people to run.

This week, you will strengthen the checks around your application, package it consistently, document how to run it, and deploy it somewhere others can inspect.

Materials:
https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/blob/main/03-deployment/lesson.md

Recording:
https://www.youtube.com/watch?v=gxt5ZDVnBMM

Companion article:
https://aishippingblog.com/p/deploy-a-full-stack-app-with-ai-coding

Homework page:
https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/blob/main/cohorts/2026/homework/03-deployment/homework.md

Submission page:
https://courses.datatalks.club/ai-dev-tools-2026/homework/hw3

Deadline: September 22, 2026 at 01:00 CET (September 21 at 23:00 UTC).

Next week, Module 4 will cover DevOps and observability for applications built with AI assistance.

Questions belong in the course Slack channel:
https://app.slack.com/client/T01ATQK62F8/C09HWT76L95

If you're looking for a way to get free or cheap AI coding assistant, I just wrote an article about it

https://aishippingblog.com/p/free-and-affordable-ai-native-development

Maybe it'll help some of you!

The homework is updated:

https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/blob/main/cohorts/2026/homework/02-development/homework.md

Have fun!

We start Module 2

This week, we move from designing an AI-native workflow to using it across a full-stack application.

Module 2 is Build and Ship an AI-Assisted Full-Stack App. The focus is the complete development loop: planning the application, coordinating frontend and backend work, testing the integration, documenting the result, and getting it deployed.

Materials:
https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/tree/main/02-development

Recording:
https://www.youtube.com/watch?v=x9dq5nBpDg8

Companion article:
https://aishippingblog.com/p/build-and-ship-a-full-stack-app-with

The current Homework 2 is DRAFT. Once it’s updated, we’ll notify you.

Submission page:
https://courses.datatalks.club/ai-dev-tools-2026/homework/hw2

Deadline: September 15, 2026 at 01:00 CET.

Next week, Module 3 will focus on testing, containerizing, and deploying the application.

Questions belong in the course Slack channel:
https://app.slack.com/client/T01ATQK62F8/C09HWT76L95

Muse Spark 1.3 is completely free in opencode

How to use it

- download and install OpenCode
- select Muse Spark 1.3 Free
- use for free while they run this campaign

Module 1 has started

This week, we begin with AI-Native Developer Workflow.

The goal is to turn a vague product idea into specified, implemented, and independently verified software with coding agents, while keeping requirements, key decisions, and final review in human hands.

You will work on:
- Turning an idea into a product specification and focused backlog.
- Giving agents durable context with AGENTS.md and project documents.
- Separating product manager, software engineer, and QA roles.
- Using loop engineering around checkable completion conditions.
- Coordinating specialized agents across a backlog with graph engineering.

Materials: https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/tree/main/01-ai-native-workflow

Recording: https://www.youtube.com/watch?v=VUJxJGpaDEs

Companion article: https://aishippingblog.com/p/ai-native-development-specifications

Homework 1 asks you to apply this workflow to a spec-driven Django application based on a deliberately vague shared-household-chores idea.

Instructions: https://github.com/DataTalksClub/ai-dev-tools-zoomcamp/blob/main/cohorts/2026/01-overview/homework.md

Submit here: https://courses.datatalks.club/ai-dev-tools-2026/homework/hw1

Deadline: September 8, 2026 at 01:00 CET (September 7 at 23:00 UTC).

The homework is useful practice, but it is not required for the certificate.
Questions belong in the course Slack channel: https://app.slack.com/client/T01ATQK62F8/C09HWT76L95

We're officially starting our AI Dev Tools Zoomcamp!

Join us to learn how to go from raw idea to production with AI coding agents

- Stream: https://www.youtube.com/watch?v=YEsrF3EX4Hw
- Article: https://aishippingblog.com/p/ai-dev-tools-zoomcamp-2026-starts
- Questions: https://qna.dtcdev.click/r/ai-dev-tools

You can also watch the stream later in recording

Have fun!

Coming back with "Use Claude Code as a Pro"!

Stream: https://www.youtube.com/watch?v=H-Ot4_1PkoM
Materials: https://github.com/thelearningdev/use-claude-code-like-pro

Join now or watch later in recording

AI Dev Tools Zoomcamp 2026 launches tomorrow.

The course teaches you how to:

- Choose the right AI tools
- Give them the right context
- Extend them with the right capabilities
- Ship safely with review, audit, security, and DevOps controls

The cohort starts tomorrow. Register now to join from the beginning and stay on track with homework, peer review, and the final project.

Register here: https://courses.datatalks.club/register/ai-dev-tools/
Join the launch: https://luma.com/tsiusx8s

Join Bhavani Ravi on Monday, August 31, for a hands-on workshop on extending and customizing Claude Code and other coding agents.

This is Workshop 3 of AI Dev Tools Zoomcamp, covering Module 3: Coding Agent Capabilities.

- Aug 31, 2:00 PM CET
- Live on YouTube

You'll learn how to:

- Use project instructions
- Connect assistants to tools and context using MCP
- Build reusable workflows and commands
- Add specialized subagents for focused work
- Design hooks for checks and automation
- Package capabilities with plugins and extensions

Register here: https://luma.com/ap4l3qlj

We published the fourth article in our AI Dev Tools Zoomcamp series, continuing the system design interview app from the previous modules.

Alexey takes the deployed application closer to production and shows how to use AI coding assistants to:
- Separate development and production environments and safely promote releases
- Add observability with OpenTelemetry, Prometheus, Loki, Tempo, and Grafana
- Define application metrics, dashboards, and alerts
- Use an AI coding agent as an on-call responder that investigates and fixes failures

Read it here: https://aishippingblog.com/p/devops-and-observability-for-an-ai

Live Q&A about AI Dev Tools Zoomcamp is today!

Join Alexey Grigorev at 5 PM CET to learn about the course structure, modules, homework, projects, and certificates.

Join if you're considering the course and want to know:

- Who it's for and what you need before joining
- What you don't need: prior AI tools experience, web development experience, a powerful machine, or a GPU
- How the four modules are structured
- How homework, deadlines, peer review, and the final project work
- Where to get help during the cohort

The live cohort starts on August 31.

Register for the course: https://courses.datatalks.club/register/ai-dev-tools/

Join the Q&A today: https://luma.com/a8qa5s2s

https://www.reddit.com/r/singularity/comments/1vu87p5/a_stealth_model_called_oxalpha_has_been_released/

If you didn't see it yet, there's a new free model with unlimited usage for 7 days (5 more remaining)

I've been running it since morning and it's pretty good

Give it a try too

We're starting the stream on devops and observability

Stream: https://www.youtube.com/watch?v=YkxLo_FRoQw
Article: https://alexeyondata.substack.com/p/devops-and-observability-for-an-ai

Join now or watch later in recording

3 events are left on our calendar before the course starts.

Join them to explore the course topics and prepare for the new cohort.

• August 18, 2:00 PM CET: DevOps and Observability for AI-Built Apps
• August 24, 5:00 PM CET: Pre-Course Live Q&A
• August 31, 5:00 PM CET: AI Dev Tools Zoomcamp Course Launch