The work
You will lead and mentor a pod of about 5 to 10 technical trainers building and reviewing realistic terminal-based tasks for AI training and evaluation. Your work combines technical quality review with hands-on checks of code, simulations, computational environments, automated tests, and task specifications.
You will review work across computer science and electrical engineering, including scientific software, distributed systems, compilers, operating systems, databases, GPU computing, circuit design, signal processing, power systems, and embedded systems.
- Review problem statements, codebases, circuit netlists, hardware description language files, datasets, reference solutions, environments, and automated tests.
- Check algorithm complexity, numerical accuracy, memory management, concurrency, circuit laws, power balance, timing, system stability, and reproducibility.
- Find nondeterminism, race conditions, unreliable thresholds, simulation convergence problems, incorrect tolerances, flawed assumptions, and ways evaluations could be bypassed.
- Review automated graders for functional correctness, performance, frequency response, waveforms, timing, efficiency, and system behavior.
- Give precise feedback, track revisions, onboard trainers, assign work, monitor throughput, and maintain technical documentation.
What it pays and takes
This is a contractor assignment listed as part-time. Pay details are not provided in the listing.
- Contract duration: 4 weeks.
- Time requirement: At least 4 hours per day and a minimum of 40 hours per week.
- Schedule requirement: Include 4 hours of overlap with Pacific Standard Time.
- Listed experience level: Entry level.
- Education or experience: Ph.D., postdoctoral experience, or equivalent advanced technical experience in computer science, electrical engineering, computer engineering, or a related field.
- Programming: Strong ability in one or more of C/C++, Rust, Python, Julia, Fortran, MATLAB, Octave, or hardware description languages.
- Technical skills: Advanced Linux and terminal proficiency, plus strong knowledge of algorithms, computational systems, numerical methods, software testing, or electrical engineering principles.
- Specialist expertise: At least one of parallel computing, systems programming, scientific software, circuit simulation, signal processing, power systems, or digital hardware design.
- Review experience: Experience assessing complex codebases, technical implementations, engineering simulations, or research outputs.
- Leadership: Experience mentoring, reviewing, or leading small technical teams, with strong written communication for detailed technical feedback.
- Helpful tools: MPI, OpenMP, CUDA, LLVM or GCC, CMake, gdb, Valgrind, perf, sanitizers, Slurm, Docker, Git, CI/CD, pytest, technical benchmarks, programming challenges, automated grading, AI coding agents, open-source maintenance, or LLM-generated technical solutions.
How it works
OpenTrain is a platform where people find AI training and data-labeling work, create a profile, and start applications. Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training is the human work behind modern AI systems, including reviewing code, testing model outputs, and creating reliable evaluation tasks. People with deep technical expertise help make these evaluations accurate, realistic, and useful for improving AI systems.