AI Agent Build Competition

Meetup contest

AI Agent Build Competition

A skills-building competition for Boston Kubernetes Meetup attendees. The goal isn’t to build the fanciest agent — it’s to practice a new skill, document what you learned, and publish it so others can learn too.

The challenge

Build an AI agent. It can be simple or sophisticated — what matters is that it’s yours, you can explain how it works, and you write it up clearly enough that someone else could learn from it.

Some ideas to get you started (pick one, riff on one, or bring your own):

  • An agent that monitors a Kubernetes cluster and summarizes pod health in plain English
  • A coding agent that reviews pull requests against a style guide
  • A support-ticket triage agent that tags and routes incoming issues
  • A personal automation agent — calendar wrangling, inbox triage, research summarization
  • An agent that wraps a CLI tool you use daily and adds natural-language control

If you’ve never built an agent before, this is the perfect excuse to start small and learn in public.

What to submit

Submit as many of the following as apply — more formats means more reach for your work, but a single solid GitHub repo is a completely valid entry on its own:

  • GitHub repo — public, with a README that explains what the agent does and how to run it
  • Blog post — published on your own site (or a platform like Dev.to/Medium) walking through what you built and what you learned
  • Your agent’s own website — if you feel so inclined, have your agent self-publish it’s own website
  • Any combination of the above

There’s no minimum length requirement. A focused, honest writeup of a small agent beats a sprawling README for something half-finished.

Documentation guidelines

Good documentation is the actual point of this competition. At minimum, try to cover:

  1. What the agent does — the problem it solves and who it’s for
  2. How it works — architecture, the model/framework you used, and any notable design decisions
  3. What you learned — the new skill or tool you practiced, including what didn’t work
  4. How to run it — setup steps clear enough for a stranger to follow
  5. What’s next — known limitations or ideas for a v2

Prize

One winner will receive the Ardan Labs Mastery Bundle — a full set of online training courses, valued at $999.

Judging

Submissions will be judged on:

  • Learning demonstrated — did you genuinely stretch into a new skill or tool?
  • Documentation quality — could someone else follow your writeup and learn from it?
  • Clarity of the agent’s purpose — is it clear what problem it solves and why?
  • Execution — does it work as described?

Polish is welcome but isn’t the bar — a clearly documented learning journey will beat a slick demo with no explanation every time.

How to enter

  1. Build your agent
  2. Publish your documentation (repo, blog post, or both)
  3. Submit your link(s) at the meetup or send them to me directly

Timeline

Submit your entry before July 16, 2026.

Questions about the competition? Reach out, or ask at the next Boston Kubernetes Meetup.