tyeetaleAbout
How I think about making things. Principles that stay, interests that evolve.
Principles
Ship, then refine
A working thing in someone's hands teaches more than a perfect thing in your head. Get it out, watch how it breaks, fix what matters.
Systems over features
Features solve one problem. Systems solve categories of problems. I always ask: what's the system that makes this feature trivial to build?
AI is an interface problem
Models are commoditizing. The hard part is designing the interface that makes AI output useful. A great model with bad UX is just a chatbot.
Data compounds
Every data pipeline you build, every connection you surface, every pattern you capture. It compounds. Build the infrastructure to learn from your own systems.
Cross-discipline advantage
The best products happen at intersections. Finance informs how I think about data. Design informs how I think about systems. Engineering makes it all real.
Intent over action
Understanding what someone is trying to do matters more than what they asked for. Map intent first, then figure out the action.
Currently Exploring
Agentic systems
How do you build AI agents that reliably act on behalf of users? What does trust look like when the agent holds your wallet?
Intent mapping
Going from 'what the user said' to 'what they actually need' to 'what action to take.' The gap between these three is where most AI products fail.
Data infrastructure at human scale
Most data pipelines are built for dashboards no one reads. How do you build systems where the data surfaces insights at the moment they matter?
AI-native UX patterns
We're still designing AI products with forms and buttons. What does interaction look like when the system can anticipate, suggest, and act?
Cost of intelligence
Running AI at scale is expensive. How do you architect systems where the right model runs at the right time for the right cost?