The work
You will create custom training environments that test whether an AI system can solve difficult competitive programming problems. The work includes designing original problems, building evaluation materials, and checking submitted solutions against the requirements.
You will also help refine problem specifications, resolve unclear cases, and explain technical decisions clearly to the project team.
- Design and implement original competitive programming problems that are not found on existing online judges.
- Create reference solutions and comprehensive test suites for evaluating AI-generated submissions.
- Develop robust C++ checkers for problems with multiple valid outputs.
- Evaluate and validate submissions against the problem requirements.
- Document checker logic and communicate technical details clearly.
- Work with the project team to refine specifications and clarify ambiguous scenarios.
What it pays and takes
- $45 to $65 per hour equivalent, paid per task that meets the project specifications.
- Less than 20 hours per week.
- Open to candidates in: AE, AR, AT, BD, BE, BH, BR, CA, CH, CL, CO, DE, DK, DZ, EG, ES, FR, GB, GH, GR, ID, IN, IT, JO, JP, KE, KG, KW.
- Open to candidates in: KZ, LB, LK, MA, MX, MY, NG, NL, OM, PE, PH, PK, PS, PT, QA, SA, SG, TH, TN, TW, US, UZ, VN, ZA.
- Professional fluency in English.
- Expertise in competitive programming and coding problem analysis.
- Advanced proficiency in C++ for building reliable checkers.
- Strong analytical skills for complex constraints and problems with multiple valid solution paths.
- Strong written and verbal communication skills, careful code validation, and close attention to output consistency.
- Ability to work independently in a remote, collaborative setting and produce well-documented code under tight deadlines.
- Experience as a problem setter, tester, or checker is helpful, as is familiarity with platforms such as Codeforces.
- IOI medals or other competitive programming achievements are helpful but not required.
How it works
You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.
About AI training work
AI training work uses examples, evaluations, and human feedback to help AI systems produce better results. People with strong technical skills are needed to create accurate coding tasks and judge whether model outputs meet precise requirements.