About Me

illustrations illustrations illustrations illustrations illustrations illustrations

I started as a control engineer. Control was the subject I did best in as an undergraduate, and it took me to CERN, where I worked on accelerator control and the programs that ran it. Two years of keeping a machine like that running taught me the habit that still defines my work: a system counts only when it works every day, not when it works once.

Watching AlphaGo beat Lee Sedol while I was there changed what I wanted to build. I went to graduate school for artificial intelligence and spent it on making deep learning efficient: model compression for object detection and segmentation, and inference on hardware with tight compute budgets. The question behind that research was practical, not academic. Most of what works in a lab never reaches a product, and I wanted to be the person who closes that gap.

That is the work I have done since. As CTO of two startups I led computer vision, on-device AI, and robotic locomotion projects, and I built and launched web and mobile products on my own, from backend and APIs to iOS and Android apps. Across all of it the pattern is the same: take an ambiguous problem, find the layer where it actually lives, and ship something that holds up in production.

Today I am the founder and CEO of TRAIL Labs, where I turn frontier models into production systems with measurable results. As the sole engineer I designed, built, and launched two SaaS products, TRAIL Search and TRAIL Studio, from first prototype to production: architecture, hands-on Python and TypeScript development, multi-agent LLM pipelines, evaluation harnesses, integrations, and operations across reliability, latency, and cost.

TRAIL Search measures how a brand is found and cited by AI answer engines and prescribes what to change. TRAIL Studio is a brand-feature-space content generation engine that produces articles, short-form video, card news, product photos, and detail pages from a single brand specification.

What I enjoy most is meeting a problem in an unfamiliar environment and solving it from many angles by drawing on everything I have built before.

Dohyeong Kim
signature