What I'm Building

Right now, two products have most of my attention — and they're really two answers to the same question: what should software feel like when it understands you?

Clara is a personal AI assistant for managing your digital life through conversation. Not a chatbot bolted onto an app — the conversation isthe interface. Calendar, messages, tasks, decisions: you talk, it handles. The hard part isn't the AI; it's the trust. Clara has to be right enough, fast enough, and quiet enough that you stop thinking about it.

Astrais the tool I wished existed every time I joined a new codebase: a conversational CLI that builds a persistent, semantic memory of your code and lets you reshape it by talking to it. Refactors that took afternoons happen in minutes. It's early, but every real-world run convinces me further.

Alongside both, BotHive keeps growing — and everything I learn shipping agents there feeds straight back into Clara and Astra.

I am also exploring what happens when a portfolio can explain the work behind it. A project should not be reduced to a logo, a tech stack, or a list of links. The decisions, iterations, and shipped details are the proof, and I want to make that proof easier to collect and share.

The thread connecting all of it: conversation as the next interface layer. I want to be early to that, and I'd rather build it than wait for it.

AI is an interface problem

A model can be capable and still make a weak product. The useful work is deciding what it should know, when it should act, how it shows uncertainty, and when it should get out of the way. That is the part I am deliberately practicing across Clara, Astra, and BotHive.