Collinear builds high-fidelity environments, tasks, and evaluation infrastructure that frontier AI teams use to understand and improve their agents.

About the role

As a Member of Technical Staff - Product, you will help decide what Collinear builds, and then build it. You’ll work directly with researchers and engineers at cutting-edge AI labs, identify their hardest problems, and turn those problems into products that advance how frontier agents are trained and evaluated.

This is a high-ownership, hands-on role at the intersection of Product, Engineering, and Research. You’ll shape Collinear’s product roadmap, prototype solutions, contribute code, run experiments, and work with the team to take ideas from ambiguous customer need to a product used in real research workflows.

What you’ll do

Set the product strategy and internal roadmap so the team ships the environments, tasks, and evaluation data that address the hardest open problems at frontier labs

Work directly with researchers and engineers at leading AI labs, becoming deeply familiar with how they build, evaluate, and improve agents

Own important technical product areas such as agent environments, eval and quality systems, etc., from initial discovery through launch and iteration

Turn ambiguous research and customer needs into clear product bets, requirements, prototypes, and measurable outcomes

Drive projects from discovery through delivery. Coordinate implementation across research and engineering teams while contributing directly through prototypes, code changes, tests, technical reviews, and debugging when that is the fastest path forward.

Keep customers and cross-functional teams aligned through clear roadmaps, demos, and playbooks, communicating progress, tradeoffs, risks, and decisions clearly

About you

You have strong product judgment and can create clarity from loosely defined technical problems

You are technically fluent enough to navigate unfamiliar codebases, understand system architecture and data flows, debug failures, review pull requests, and make focused implementation changes

You have worked on developer tools, AI infrastructure, evaluation systems, data platforms, workflow products, or similarly technical products

You can reason carefully about quality: what should be measured, what evidence proves something works

You communicate effectively with researchers, engineers, customers, and company leadership, adjusting the level of technical detail without losing precision

You operate well in a fast-moving environment where the roadmap evolves as new customer and model evidence emerges

You are comfortable being accountable for the outcome even when you are not the sole implementer

You are an unusually effective user of AI development tools and know that generated work still requires strong scoping, review, testing, and judgment

Nice to have

Experience with AI agents, evaluations, benchmarks, reinforcement-learning environments, task generation, or verifier design.

Familiarity with computer-use or MCP-based agent environments.

Experience designing quality or release systems for technically complex customer deliverables.

Experience working directly with frontier AI labs, applied research teams, or enterprise technical customers.

Early-stage startup experience spanning product, engineering, operations, and customer delivery.