Now

I spent the last year building around agents — a coding agent, a context-aware meeting notepad, an AI growth teammate. Along the way, two questions started pulling at me.

What's still missing for agents to be truly capable? Context, real-world interfaces, ways to verify their work — the gaps keep revealing themselves the more you build. I want to keep pulling on these threads.

Once they are capable — how do they solve problems we can't? Agents have properties that are genuinely superhuman. They run dozens of experiments in parallel. They operate without ego. They never burn out, never need to be scheduled — just capability that can be brought into your life anywhere, anytime. And these are just a few. Each one unlocks possibilities we haven't fully explored yet. The thought exercise I keep coming back to isn't how to automate what we already do, but what becomes possible when something this capable approaches problems we cannot solve today.

Let's talk

I'm looking for what's next. If any of this resonates, I'd love to talk.