I build what I'd want to use.

I'm Peyton Lambourne — a 23-year-old developer from West Jordan, Utah, just outside Salt Lake City. I work at the intersection of AI tooling, web development, and creative technology.

Right now I'm deep in local LLM tooling, agent systems, and AI red-teaming. I built CanIRunAI to help people find the right open-source model for their hardware, WrapsRL to generate AI-powered Rocket League car wraps, and Adversec to automate adversarial testing for LLM deployments.

I'm self-taught. I started writing code years before LLMs hit the mainstream — building Minecraft bots that could carry a natural-sounding conversation in chat, back when that meant rolling your own heuristics and state machines instead of calling an API. That side project turned into a long detour through web development and eventually into AI, where most of my focus sits now. In 2025 I completed the University of Utah's Machine Learning Bootcamp and earned their Artificial Intelligence certificate.

Outside of software I make music as Lambourne — a rapper, producer, and singer/songwriter out of Salt Lake City with a catalog built around emotional, introspective hip-hop and trap. I haven't released publicly in about two years; instead I've put everything into UNVAULTED, an app I built that hosts all of my unreleased songs, snippets, and demos in one place. I also run BTRFLY Tournaments, a Rocket League esports org I founded and still play in — we've been hosting live 1v1 tournaments for the past two years.

Day-to-day I work as a barista at Starbucks. Music stays a hobby, and the day job keeps the indie projects self-funded.

My rule is simple: build what I'd actually use, and make sure it looks the way I'd want to use it. Visual polish is usually what decides whether I pursue a project at all — if the interface doesn't feel right, neither does the product. Most of my stack is Next.js, TypeScript, Supabase, and Python, deployed on Vercel.