Model Test and Measurement Engineer

About the role

Own the evaluation framework that verifies AI systems perform as intended, producing objective evidence that informs release decisions and builds customer trust.

What you will do

• Create labeled ground truth datasets and design statistically sound audit and sampling methods
• Gate model releases by version and keep evaluation records, version inventories and rollback triggers
• Run drift and override reviews and produce human oversight and fairness evidence
• Measure workflow gains such as review time, throughput, backlog movement and rework rates
• Keep independence from build roles, producing evidence without approving thresholds or releases

What they look for

• At least five years of model validation as a named responsibility rather than a side task
• Hands-on experience constructing ground truth and designing sampling
• Strong Python and SQL skills
• Ability to design evaluation phase quality control and fair comparisons across differing methods