AI measurement and experimentation

Overview

ModelMetrics is my AI model measurement and experimentation project. It sits alongside my broader interest in understanding where AI tools help with engineering work and what evidence should inform their use.

Motivation

Choosing an AI tool is an engineering decision. I am interested in examining its behavior in the context of the work it is expected to do, rather than treating model capability as a substitute for requirements and judgment.

Current status

The implementation repository is private as of October 23, 2026. This public overview stays at the project-purpose level. It does not present benchmark results, model rankings, or a released tool.