Hi, I’m Dave
I’m an AI consultant. I build and validate AI systems.
My focus is on the gap between “it works in the demo” and “I’d stake my name on this in production.” That gap is where most AI systems live, and closing it — build it, measure it, prove it, whatever it takes — is the work I find most interesting.
What I do
- Build the reliability layer — labeled eval sets, accuracy and failure reports, confidence scoring, audit logs, PII redaction. Retrofitted onto an existing pipeline, or built in from day one.
- Validate and benchmark models — does a frontier model or a specialized OCR API fit your use case? I’ll answer that with numbers, not opinions.
The common thread: I care about what happens after the demo. Reliability, auditability, and security are the defaults — not the add-ons you bolt on when something breaks. My deepest proof of this so far is in document and financial AI extraction, but the same rigor applies wherever an AI system touches data you can’t afford to get wrong.
Get in touch
If your AI system needs a reliability layer — someone to build it, or to prove it works — reach out.