Board Game Reasoning Expert (AI Training & Evaluation)
- Game Development
Listed through Turing.
As a Software Engineering evaluator, you will create cutting-edge datasets for training, benchmarking, and advancing large language models, collaborating closely with researchers. This includes curating code examples, providing precise solutions, and making corrections in Python, JavaScript (including ReactJS), C/C++,…
Not disclosed
Turing screens per job, not platform-wide, and most jobs here end in a live 60-minute technical interview with a person, usually followed by a short cultural and offer conversation. Others replace that with a timed assessment, a work-sample review, or — on one job — an AI video interview. The posting's own "Evaluation Process" section is authoritative; read it before you apply.
Turing matches you to a posting rather than a general pool, so the steps below depend on which job you picked.
Your profile and CV are screened against the job's qualifications and relevant professional experience.
Most commonly a 60-minute live technical interview with an engineer, sometimes with live coding. Some jobs swap this for a timed assessment due within 24 hours of being sent, an automated coding challenge, a take-home, or a short work sample — hands-on and language jobs often ask for a delivery review or a 30-second demo recording.
Where there's a second round it's short — 15 to 30 minutes covering fit, the offer and working conditions.
Nearly all of these are contractor assignments with a fixed duration and no paid or medical leave. Hours per week and required timezone overlap are set at this point.