Session three of the Hitchhiker’s Guide series — agents, tools and workflows, read and debated together.
What we covered
Small room, deep discussion. Session three took on the workflow question: how do agent steps compose into reliable work?
- Single-agent vs. multi-agent — when one loop suffices and when to split roles.
- Handoffs and supervision — patterns for agents checking each other’s work.
- Workflows as code — where the graph is explicit, where it is emergent, and how to test either.
What we built
A whiteboard of workflow patterns argued over by 16 regulars — the kind of session where everyone talks and the notes write themselves.
Takeaways
- Small sessions go deepest: 16 people, zero lurkers, every voice in the mix.
- Multi-agent is an organizational pattern first and a technical one second.
- If you cannot draw the workflow on a whiteboard, you cannot debug it in production.
Join the next one
We run sessions like this regularly — free, welcoming, and open to everyone. RSVP on Meetup and bring your questions.
