selected work
Five projects, zero tutorials followed. Each one taught me something the hard way.
01 — Runtime control for LLM applications
A guardrail layer that wraps the OpenAI client — per-session budget reservations, tool allowlists, execution policy. No proxy, keys never leave the host app. Span-first tracing with violation capture and eval-ready dataset export.
02 — Multi-model AI image editor
20+ LoRA style-transfer models, AI background removal, and a canvas-based multi-layer editor. Launched on Product Hunt, monetized with tiered subscriptions, and publicly praised by Vercel's CEO.
03 — Phone automation, fully local
End-to-end Android automation inspired by OpenClaw — an agent that drives your phone, running entirely on-device. No cloud, no leaks, just a phone doing what you tell it.
04 — A vector database from scratch
Custom vector DB built in Elixir to explore indexing architectures — HNSW, IVFFlat — from first principles. Because the best way to understand a database is to build one.
05 — A browser that lives in your terminal
Terminal-based browser engineered in Go, driving Brave in headless mode. The web, rendered where developers actually live.
0+
requests/day my APIs handle without crying
0+
LoRA models wired into one image platform
0
peak LeetCode rating. top 0.3% worldwide
0%
manual work erased by agent workflows I shipped