AI · Project
Evebot — An AI Assistant Experiment
A hands-on AI assistant experiment — building to understand where the technology genuinely helps, and where it only looks like it does.
The context
It's easy to have opinions about AI from the outside. It's harder — and far more useful — to build with it. Evebot started as a way to close that gap: a personal assistant experiment, built to learn by doing.
What I built and why
The point wasn't a polished product; it was fluency. Working directly with modern AI tooling to feel where it's genuinely capable, where it's brittle, and where a strategist's judgment still has to sit in the loop.
That firsthand sense now informs how I think about AI in a fintech and Web3 context — separating what's real from what's narrative.
How I approached it
Build the smallest thing that tests a real question. Use it honestly enough to hit the rough edges. Treat every failure as information about where the technology's true boundary sits today.
What it added up to
A grounded, non-hype view of applied AI — the kind you can only earn by building — and a habit of forming technology opinions from evidence rather than headlines.