SF October 14th: A Birds of a Feather Session on Agentic Engineering
Simon Willison’s Weblog Simon Willison
Simon Willison and Jesse Vincent are hosting a San Francisco meetup for people building with coding agents. It’s for odd, unfinished ideas — not sales pitches.
Based on reporting by Simon Willison’s Weblog, Simon Willison — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Simon Willison is putting together an evening session in San Francisco on Wednesday, 14 October with Jesse Vincent, and the focus is narrow in a good way: people building weird and interesting things with coding agents. Not polished demos. Not product launches. More like a room full of builders comparing notes on what they’re trying, what they’ve learned, and what still makes no sense.
The event is being framed as an agentic show-and-tell, but the vibe sounds looser than that. People are encouraged to bring along work they haven’t talked about publicly, strange experiments, or projects that are still too unfinished to have an obvious market. If someone wants to share what they’re working on, great. If not, that’s fine too.
There’s no requirement to present anything, which is probably the point. The evening is meant to be one flowing conversation, with informal show-and-tell folded in rather than staged on top. That should make it easier for the kind of people who are tinkering in the gaps between useful and ridiculous — where a lot of the interesting agent work seems to live right now.
And the announcement leans into that directly. This isn’t being sold as a neat, mature category with tidy answers. It’s a nod to the fact that agentic AI is still strange, still messy, and still full of half-built ideas worth discussing before they get packaged into something more respectable.
My take — AI-written commentary, not fact-checked reporting
This is the right way to handle coding agents right now: keep the pitch decks at home and talk about the messy parts out loud. The sector has already seen enough polished demos pretending to be finished products. A room full of unfinished experiments is usually where the actual signal is hiding.
Read more about this at: Simon Willison’s Weblog