About
AI engineer building products and systems people can actually use.
I work across AI products, data systems, and software — the common thread is trying to make things clear enough that people trust them.
Opening
I’m Tim Finch. I build AI products and the systems behind them, and I try to keep things simple.
Current focus
My work is a mix of AI engineering, product work, and data infrastructure — using AI to explore and build new products, and building the systems that make them work.
Background
I love to work on LLM applications, agents, APIs, automation services, and data pipelines, built with contracts I can point to and behavior I can actually measure.
Working style
I love writing about what I’m building and sharing it as I go — it’s a good way to learn in public, and I’d like to think it’s useful to whoever’s reading too.
Current build
CityScout
CityScout is built around a simple idea: travel feels better when the product understands the city, not just the booking. The iOS app keeps trip state close to the device, while a FastAPI backend handles itinerary generation, guide chat, and structured AI outputs. A future web layer will support planning and sharing, but iOS remains the in-trip companion.
Principles
- Ship fewer things, and make them easy to understand.
- Keep boring, predictable architecture in the critical path.
- Test and evaluate as you build, not after.
- Use design to cut clutter, not add polish.
Working record
I write about what I’m building as I go, mostly so I don’t lose track of why I made a decision once the product’s moved on from it.