model evaluation · agent training · procurement workflows
ABOUT THE ROLE
Join an early-stage AI team building agents for manufacturing operations. You will create benchmarks and train models against real procurement and supply-chain workflows, helping ensure agents can handle customer-facing tasks reliably.
WHAT YOU'LL DO
- Design and build evaluation suites for supplier sourcing, qualification, compliance documents, purchase orders, freight, and order tracking.
- Work with procurement and manufacturing experts to scope tasks, write reference answers and rubrics, and build graders.
- Train and fine-tune models, using evaluation results to guide experiments.
- Identify agent failures, improve production systems, and measure whether changes work.
- Help establish the team's research and evaluation practices.
WHAT WE'RE LOOKING FOR
- At least 3 years of experience, operating at a senior or staff level.
- Hands-on experience designing benchmarks and evaluations, including authoring tasks, reference solutions or rubrics, and quality checks.
- Practical experience training and fine-tuning models, not only using models through APIs.
- Ability to work closely with non-ML domain experts and translate their knowledge into measurable tasks.
- Supply-chain, procurement, manufacturing, or early-stage startup experience is helpful.
COMPENSATION & BENEFITS
Base salary is $180,000 to $250,000 annually, plus equity. Visa sponsorship details were not provided.
LOCATION
Hybrid in San Francisco, California, United States.