Careers · 1 open role

We are hiring.

Strust is building the validation loop for code migrations. We capture what a source system does, validate every conversion, turn mismatches into stronger checks, and carry that evidence into the next migration wave. The eval harness makes that loop real. You would own it.

Open role

Member of Technical Staff, Evals

Remote

Apply

About the work

The harness runs paired executions of source and target systems under controlled data, captures traces of everything that matters — branches, rounding, ordering, side effects — and scores equivalence. Today it goes deepest on COBOL→Java. It needs to get faster, broader, and harder to fool.

This is infrastructure work with a research edge: the interesting problems are what to measure, how to measure it defensibly, and how to run it at a scale where the numbers hold up in front of an auditor.

What you'll do

Build the validation loop for code migrations, so every mismatch becomes a stronger check for the next conversion.
Own the eval harness end to end: trace capture, differential execution, equivalence scoring, and the evidence reports customers sign off on.
Design evaluations that measure behavioral parity between source and target systems — and defend why each metric is the right one.
Build the infrastructure to run tens of thousands of paired executions reliably, reproducibly, and at scale.
Turn known failure modes — rounding modes, hidden branches, date rollover, job scheduling, side effects — into detectors that catch them automatically.
Extend the harness beyond COBOL→Java toward any→any language coverage.
Work directly with the founder. Your output goes in front of customers as signed evidence packs.

You might be a fit if

You write strong Python (JVM familiarity helps) and have built production or research infrastructure that other people depended on.
You have operated distributed systems, data pipelines, or test infrastructure that had to be reliable at scale.
You care about correctness at the level of rounding modes and branch semantics, not just green CI.
You find differential testing, metamorphic testing, program analysis, or eval harnesses genuinely interesting.
You communicate clearly in writing. Our reports are read by executives and auditors, not just engineers.

Nice to have

COBOL, mainframe, or legacy-modernization exposure.
Compilers, static analysis, or symbolic execution background.
Experience building LLM eval harnesses or LLM-assisted code translation tooling.

What you get

Founding-engineer equity and a say in everything.
One concrete mission: make the harness impossible to argue with.
Direct customer exposure — the evidence you produce is the product.

How to apply

Email [email protected] with a short note and a link to something you have built. That beats a résumé.

Apply for this roleBack to home