Hiring for eval infrastructure

Build the evaluator agents cannot grade.

We are looking for engineers who want to make agent-written software easier to test and behavior changes easier to understand.

Role surface

Open roleMember of Technical Staff, EvalsRemote / Full-time

BuildDeterministic runtime

IsolateFirst divergence

ReturnReviewable evidence

Current opening

Own the evaluator end to end.

This role spans runtime control, differential execution, agent feedback, and evidence a human can challenge.

Open roles

Member of Technical Staff, Evals

Remote / Full-time

Build the deterministic runtime and evaluation architecture behind Strust's behavioral feedback loop.

How we work

We value people who can move between careful technical reasoning and shipping useful work. You do not need to match every line of a role. Show us what you have built, what you learned, and how you think about the details.

Loading page…