Kenneth W. Bingham

I am a software engineer with more than ten years building reliable systems end to end, across regulated, high-stakes domains: U.S. Air Force programs (I held a Secret clearance), healthcare and insurance, fintech, and fleet payments. I work full-stack, from database and API to a modern front end and the cloud deploy that puts it in production.

Since August 2025, on my own initiative, I have gone deep into AI engineering: not only how to use large language models, but how they work and how to make them reliable through operating directives and geometric representation. I build the reliability layer around AI: deterministic pipelines, evaluation and verification harnesses, and reproducible systems. I prove my work with tests and reproducible numbers, and I separate what is established from what is a model from what is still a conjecture under test, so the bold ideas stay credible.

• Determinism over fluency. The model is nondeterministic; the system around it must not be. Schema-validated outputs, retry budgets, and verifier passes, so malformed or drifting output never reaches the next stage.
• Cost where it counts. Route cheap calls to cheap models, cache byte-identical context, budget tokens, so frontier spend goes only where frontier reasoning is required.
• Reproducibility and provenance. Content-addressed, hash-verified, deterministic replay, so you can prove exactly what a system did and recreate it on any machine.
• Ownership end to end. I run my own Linux infrastructure with zero-downtime, auto-rollback deploys, and I own problems from architecture through production.

I am a drummer, a singer, and an acrylic painter, and an independent researcher in geometric, representation-based computation. The same instinct runs through all of it: find the structure underneath, and let it carry the work.