How to Build Cool $#!+ (That Actually Works)
A crash course in agentic engineering
Anyone can now describe an app in plain language and have something running by the afternoon. Keeping it running is a different skill, and it’s the one this course is about.
Agentic engineering is what comes after vibe coding. You think the problem through, give the agent what it needs to make the same calls you would, and build the checks that tell you whether it worked. We go from a blank page to a small working project, and spend most of the time on the parts a demo skips: the spec, the verification, and what happens when it breaks.
Who it’s for
For people who want to build with AI agents and have it hold up: thinking it through, doing it, and making it robust.
What we cover
- 01
Thinking it through
What am I actually building?
- 02
Doing it
How do I work with an agent?
- 03
Making it robust
How do I know it works?
- 04
Living with it
What happens after it works?
What you’ll leave with
- A working project
- Something small and real, built with an agent, that you can explain
- A spec template
- A way to tell an agent what you want, what can’t break, and what done looks like
- A verification habit
- A checklist for proving it works before you believe it
- Your own setup
- An agent, a project and version control, working on your machine
Booking
For dates, format, delivery arrangements and price, contact Noam about this workshop.
