Bring one clear problem.
Tell me what your AI system may do, where human authority must remain, or which claim needs evidence. I read every message and respond personally with whether I can help and the smallest sensible next step.
Best fit: full-time roles, research collaboration, and selective advisory or consulting work involving agent evaluation, governance controls, auditability, or human oversight. Based in San Diego; include location and working-model requirements in your note.
Message sent
Thanks — it's on its way to my inbox and I'll reply to the email you gave. If it's urgent, you can also email me directly at tom@tombudd.com.
github.com/tombudd
The public artifact layer behind the work.
in/tom-budd
Professional background and history.
See the portfolio
Benchmarks, audit templates, and research.
What happens next. I will respond with one of three answers: I can help and suggest a small next step; I need one or two clarifying details; or I am not the right fit and will say so directly. No oversized discovery process is required to start.