Philosophy

How I build

Start with the problem

Don't build a feature because it sounds impressive. Understand the user, workflow, and actual pain first.

Make complexity feel simple

The best products can hide complexity without hiding important information.

Build, test, learn

I like getting ideas into users' hands quickly, learning from behavior, and improving the product.

Optimize for usability

A technically powerful product is useless if people don't understand how to use it.

My favorite feature is usually the one that removes three other features.

AI + Product

Where AI fits

I treat AI as a product capability, not a checkbox. That means deciding where a model genuinely improves a workflow, grounding it in the right knowledge, designing for trust, and knowing when the product should hand the user to a person instead.

I care less about putting AI everywhere and more about finding the places where it genuinely makes a product faster, smarter, or easier to use.

GPT
AI support experiences
Multilingual AI
AI agents
RAG / knowledge-grounded experiences
Guardrails
Hallucination reduction
AI workflow automation
Risk & fraud detection
API integrations