The longer story.

• Ship the smallest useful thing first, then iterate. Both Tetherlab and Daredate started as scrappy MVPs that earned the right to keep existing through real users, not pitch decks.

• Read the source. Whether it's a paper, a framework, or someone else's code. I'd rather understand what's actually happening than reach for an abstraction that papers over it.

• Stay close to the user. For Daredate I ran the user interviews, wrote the copy, did the marketing. For VeraConnect I sat with the growth team and watched them use the pipeline I'd built before declaring it done.

• Write things down. Plans, decisions, post-mortems, the half-formed ideas that aren't ready to act on yet. This site exists because I'm bad at keeping things in my head.

I'm increasingly interested in the boring, load-bearing infrastructure that makes agent systems trustworthy enough to put in front of real users — the coordination layer, the audit trail, the part that turns "an agent ran" into "here is what it did and why." Tetherlab is my current attempt to contribute to that layer.

Outside of code, I've kept moving: currently the Netherlands after growing up in Romania and a year in Pennsylvania. I read more than is reasonable, write occasionally, and care about clean energy and democratic processes. I'm an introvert who has somehow ended up in roles that require a lot of talking to strangers, and I'm grateful for it.

Four working principles I try to hold to. They aren't aspirational — they're how I actually operate when I'm working with you, on something that matters and on a deadline.