About
From how people learn to how machines learn
I studied cognitive science at Rice. Now I build AI systems, and the tests that check them. I am based in Dallas-Fort Worth, and I work remotely or travel to your team.
Background
How I got here
At Rice University I studied cognitive science (B.A., 2017): how people think and learn, and how those ideas can help machines learn.
In 2019-2020 I did a nine-month AI and ML bootcamp. Then I spent about five years in business, as a solutions engineer, an account manager, and the owner of my own business.
Since late 2025 I have built AI systems full time: computer vision, AI agent platforms, and the LLM eval harnesses that prove they work.
I speak both sides. Years of sales and client work taught me to scope like a consultant. I build like an engineer.
Also experienced in
Other areas
These areas have no project pages. I can talk about the methods.
- On-chain analytics and token-safety tooling
- Smart-contract auditing and security research
- Mobile apps (Expo / React Native)
- Logistics and estate-sale client POCs
FAQ
Questions
- What do you build?
- I build AI systems that ship: computer vision models, AI agent platforms, and the LLM eval harnesses that prove they work.
- Are you available for AI consulting in Dallas-Fort Worth?
- Yes. I take fixed-scope AI consulting work. I am based in the Dallas-Fort Worth area, and I can meet clients in person there.
- Do you work remotely?
- Yes. I work remotely with teams anywhere, and I travel to work on site when it helps. Each week you see a working demo.
- What does an LLM eval harness setup include?
- You get an eval corpus from your own data, a scoring function, a CI gate that returns pass, fail or error as JSON, and a runbook. It takes two to three weeks.
- What is an AI Feature Audit?
- A three-day review of one AI feature, its data and its tests. You get a written report and a fix plan in priority order.
- What is harness engineering?
- Harness engineering is building the checks around AI work. An LLM eval harness grades each change to code, models or data and returns pass, fail or error. Error means the check could not run.
- What computer vision work have you done?
- I trained YOLO and RF-DETR detection models that find HVAC symbols on construction drawings for takeoff and estimating. My best detector reached mAP50 0.755.
- How does an engagement start?
- It starts with a 30-minute intro call. Then you get a written scope and a quote. Most engagements take one to four weeks.
- Can client details stay private?
- Yes. This site uses codenames for all client work. I can sign a mutual NDA before I see your data.
- How fast do you reply?
- I reply within two working days.