
Thoughts on building AI agents, automation, and the future of invisible technology.
Most 'AI agents' are just fancy workflows. Here's what we learned about building agents that actually make decisions and adapt.
A technical deep dive into how agents make decisions, handle failures, and adapt to new situations using structured outputs and retry logic.
What we learned building Fantomu: the technical challenges of creating reliable AI agents and how to solve them.
Why the best technology disappears into the background, and how this principle guides how we built Fantomu.
The story behind the name and what it represents about the philosophy of invisible execution.
Practical examples of how AI agents can automate common development tasks. Some work, some don't - here's what we've found.
Our take on the current landscape of AI automation tools and where things are heading. Spoiler: most tools aren't actually agents.
Sharing Fantomu publicly for the first time. It's a side project, not a startup, and that's okay. Here's what I've built so far.
Get updates when I write about AI agents, automation, and building Fantomu.
No spam, just valuable insights. Unsubscribe anytime.