Careers / Expert network
We’re building a network of engineers, ML specialists and domain experts for AI data and evaluation projects. Have at least 2 years of relevant experience? If you enjoy solving problems and explaining your work, we’d like to hear from you.
Project availability varies. We discuss the scope, working arrangement and payment terms before any assignment begins.
Author hard, domain-grounded terminal tasks that frontier AI agents fail, with oracle solutions and verifiers.
Deep expertise in a technical domain; Linux, Docker, Python or a systems language; rigorous test design.
What you could work onExpert task packages: instructions, stateful environments, oracle solutions, verifiers and difficulty notes.
Design evaluation cases and check model outputs against a clear rubric.
ML fundamentals, experiment design, Python and careful error analysis.
What you could work onEvaluation datasets, scoring criteria, result analysis and calibration notes.
Bring subject knowledge to tasks where technical correctness needs specialist review.
Demonstrable experience in your field and the ability to explain a decision clearly.
What you could work onDomain-specific tasks, reviewed answers and evidence-backed feedback.
Compare responses, record preferences and check training examples for consistency.
Careful reading, structured judgement and experience following review guidelines.
What you could work onPreference comparisons, annotations, rubric decisions and correction notes.
Make task environments repeatable and investigate setup failures.
Linux, Docker, dependency management, shell workflows and CI.
What you could work onTask environments, build scripts, verifier setup and reproducibility checks.
Review task instructions, source code and acceptance tests before delivery.
Code review, edge-case testing and clear, actionable feedback.
What you could work onQA records, defect reports, correction checks and delivery checklists.
How we bring people onto a project
Send your resume, skills and availability. We look for 2+ years of relevant experience.
We review your background against the project requirements.
Where needed, we agree a focused assessment and explain what it covers.
Sign the NDA and ICA, then confirm scope, review criteria, access and payment terms before starting.
