Full-stack engineer · Houston, Texas

I build AI features that would rather be right than impressive.

Public crime-data tools at Crime Stoppers of Houston, a job-search platform used in production, and a local-first commit tool that never phones home. Starting an M.S. in Applied Computing at the University of Oklahoma this month, and running Enoch Studio.

Now Crime Stoppers of HoustonAlso Enoch Studio · M.S. at OUOpen to interesting problems

Constraining a model toward honesty costs you the clever version. Two places I paid that, and one where I haven’t.

Resuzen: The ATS match score is computed in code, not guessed by the model. Interview prep names the gaps in your résumé instead of inventing matches for them.
Verbatim Commit: Caught a 4B model mislabeling renames as “updates,” so I traded fluency for accuracy and moved the default to qwen2.5-coder:7b. The prompt forbids invented rationale.
Fixr — still owed: The highest stakes of the three and the weakest guardrails. Nothing sits between the model and the screen — no source consulted, no claim scored. I’d rather write that down than let a case study imply otherwise.

Read the essay — including the benchmark that contradicted the claim it replaced.

Selected work

All projects →

AIProductSep 2025 – ongoing

Resuzen — Job-Search Toolkit

A LaTeX résumé builder wrapped in a full job-search toolkit. Every tracked application links back to the exact résumé version you sent. Metered AI runs on atomic, fail-closed quotas that auto-refund a failed provider call.

Next.js · React · JavaScript · Tailwind CSS · Supabase · Stripe · OpenRouter · LaTeX · PostHog · Sentry · Vercel

100+Resumes generatedRead case study

AIToolingJun 2026

Verbatim Commit — Local-First AI Commit Messages

Local-first commit messages generated from your staged diff against your own Ollama model. Diffs are budgeted head-and-tail with explicit omission markers, and the hook reads the controlling terminal so it can never block CI.

TypeScript · Node.js · Commander.js · Ollama · qwen2.5-coder:7b · Git Hooks · node:test · VHS

0Cloud callsRead case study

AIProductFeb – May 2025

Fixr - Vehicle Maintenance Tracker

Vehicle maintenance tracking with Fixy, a Gemini-backed assistant that explains service needs in plain language. One model call, rendered verbatim — a known, named gap where a verification layer should be.

Next.js · Vercel · JavaScript · Convex · Gemini API · NextAuth · Resend · Tailwind CSS

Read case study

Open to interesting engineering problems

Houston-based, open to remote. Résumé, GitHub, and case studies one click away.

Get in touch

I build AI features that would rather be right than impressive.

Now Crime Stoppers of HoustonAlso Enoch Studio · M.S. at OUOpen to interesting problems

Constraining a model toward honesty costs you the clever version. Two places I paid that, and one where I haven’t.

Resuzen

The ATS match score is computed in code, not guessed by the model. Interview prep names the gaps in your résumé instead of inventing matches for them.

Verbatim Commit

Caught a 4B model mislabeling renames as “updates,” so I traded fluency for accuracy and moved the default to qwen2.5-coder:7b. The prompt forbids invented rationale.

Fixr — still owed

The highest stakes of the three and the weakest guardrails. Nothing sits between the model and the screen — no source consulted, no claim scored. I’d rather write that down than let a case study imply otherwise.

Read the essay — including the benchmark that contradicted the claim it replaced.