About
I'm a software engineer who ships fast, mostly building developer tools around large language models. I wrote zac, a coding agent built from scratch in Zig with zero dependencies, and cdesign, which scrapes real websites into installable design specs so AI agents build on-brand UI (shipped as an npm CLI). Before that I founded 9thgen.ai, a voice-agent platform now taking calls for paying clients. At Y Lift I inherited an undocumented production app, taught myself .NET and Blazor from scratch, and shipped a full platform migration in one month for 10,000+ monthly users at 99.8% uptime. Hand me an unfamiliar codebase and I'll get something working. (I also hold multiple years of US work authorization, so sponsorship isn't a near-term concern.)
Work Experience
Lead frontend developer shipping new production features from scratch, often daily. Rebuilt the storefront (catalog, cart, checkout, auth) through a full migration from Flutter to .NET 8 + Blazor for an app serving 10,000+ monthly users at 99.8% uptime.
Founded and led an AI voice-agent startup that cut customer service costs by up to 80%. Launched on Vercel, processed 1,500+ calls in the first month, acquired 5 paying clients at an 11.6% conversion rate, and boosted customer engagement 150% through natural-language voice interactions. Owned end-to-end product development and cross-functional teams, cutting development time 40%.
Built 8 RPA automation bots with Automation Anywhere, cutting manual effort 70%. Led development of an RPA bot to extract resumes from the iCIMS platform, improving efficiency 15%, and collaborated on automated testing that raised system reliability 20%.
Education
Skills
I work across the whole stack of AI product development. On the model side I've built real-time voice pipelines, agentic tool-loops, and RAG-style workflows against the Anthropic, OpenAI, and OpenRouter APIs, and I wrote a coding agent from scratch in Zig. On the frontend I ship React, Next.js, and TypeScript interfaces. On the backend I've worked in FastAPI, .NET, and Postgres, and provisioned real cloud infrastructure with CloudFormation and the AWS SDK. The part I like most is the messy middle: taking a raw LLM capability and making it reliable enough for people to actually use.
FrontendPrimary focus
AI / LLM
Backend / Infra
GitHub Activity
Some of the stuff I've shipped
A voice platform with paying clients, a coding agent written from scratch in Zig, and a bunch of things in between. Plenty more on my GitHub.
zac
A coding agent that lives in your terminalA coding agent I wrote from scratch in Zig, about 3,800 lines, one dependency-free binary that drives Claude, GPT, and Gemini from the terminal.
9thgen.ai
Real-time AI voice agentsAI voice agents that take customer calls end to end and cut support costs up to 80%. Handled 1,500+ calls across 5 paying clients in the first month.
Orbit.build
AI-native development environmentAn environment where agents build and ship code on their own. I built the developer-facing React UI and worked on the pipelines under it, now at 200+ early users.
ctx
Get back up to speed on any codebase in secondsA CLI that scans a git repo and tells you where things stand: recent work, uncommitted changes, stray TODOs, and an AI read on what to do next.
Chrome AI Sidebar
Open-source browser extensionA privacy-first Chrome extension for instant AI on any webpage, wired to OpenRouter so you can plug in any LLM.
MindMap SaaS
Prompt-to-mind-map generatorTurns a prompt into an editable mind map you can expand and rearrange live, with Anthropic wired in to refine it through chat.
Let's build something
Hiring, collaborating, or just curious? Reach out — I'd love to hear from you.