Back to Home

What I'm building right now.

Not an archive. This is where the thesis gets tested in the open: live product work, AI governance, and experiments I run to keep my own hands on the tools.

Live

Product safety and design governance for an AI Assistant with agentic features

Designing a tiered safety governance model for an AI companion platform built on LLM chat and agent orchestration: behavioral boundaries, risk-classification tiers, and escalation protocols so the system behaves within defined and documented responsible parameters at scale.

Live

AI Product Design and Operations lead, Cairya

Owning the AI product lifecycle for an education and coaching platform across mobile, desktop, and VR. Defining audit-agent requirements that verify model behavior against approved content, and setting the evaluation criteria that keep hallucination risk down.

Building

An independent AI governance practice

Helping healthcare, workforce, and education organizations adopt AI responsibly: model behavior boundaries, audit approaches, and the change management that adoption actually depends on. Pursuing the AIGP credential, with ISO 42001 to follow, ahead of EU AI Act enforcement.

Experiments & prototypes

Hands on the tools.

I don't advise and direct on AI or design from the sidelines. These are things I built myself to stay close to the craft. Rolling up my sleeves is one of the most profound ways to empathize with the teams I manage. There's a time to delegate, and time to dive in with the team. Leadership is knowing when.

Sketch to startup: the framework, working

An interactive demo of the framework that runs an idea from rough sketch toward a startable product. Open the live demo →

Google AI Studio prototypes

Working prototypes exploring AI-assisted product flows.

Previews coming soon!

Three screen designs built in Claude Code

End-to-end screen designs taken from prompt to working front-end.

Previews coming soon!