12 hours ago
Remote, United StatesSenior
Responsibilities
- Write original competitive-programming problems with clear statements, sound constraints, and intended solutions.
- Implement correct solutions in C++ and Python and create realistic incorrect solutions representing common mistakes.
- Build generators, edge cases, stress tests, and hacking tests to expose flawed implementations.
- Write testlib checkers and interactors for problems with multiple valid answers.
- Verify that correct solutions pass and incorrect solutions fail within the required time limits.
- Contribute to testing, evaluating, and improving AI coding models through project-based tasks.
Requirements
- Strong competitive-programming background demonstrated by a public Codeforces profile or IOI, ICPC, or national-olympiad achievement.
- Fluent contest C++ skills, including STL and complexity analysis, plus comfort writing Python.
- Ability to explain why solutions are incorrect and construct inputs that break them.
- Strong written English at the C1+ level.
- Applicants should submit competitive-programming handles and ratings, problem-setting experience, and IOI/ICPC/olympiad history.
- Testlib experience, real contest problem-setting or testing experience, and LLM code-benchmark or evaluation-dataset experience are preferred.
Benefits
- Remote, part-time, project-based freelance work rather than permanent employment.
- Estimated workload of approximately 10–20 hours per week during active project phases, with no guaranteed workload.
- Paid per accepted task, with an effective rate of up to the equivalent of $90 per hour depending on qualification tier and efficiency.
- Tasks are completed after qualification and project assignment and must meet stated deadlines and acceptance criteria.
About Mindrift
Mindrift builds an expert-sourcing platform that connects domain specialists to project-based work training and evaluating generative AI models, including supervised fine-tuning, RLHF, evaluation, and red-teaming. It is built and operated by Toloka, part of Nebius Group, and run from Amsterdam, Netherlands. Work is fully remote and freelance, serving global technology companies developing and improving large AI systems.