The work
This contractor role focuses on creating realistic machine learning engineering problems and evaluating how advanced AI systems solve them. Day to day, you may write task instructions, assemble reproducible environments, build reference solutions, define objective grading criteria, and inspect generated code, experiments, analyses, or other technical deliverables.
What the screening emphasizes
micro1 is seeking candidates with prior paid experience training, evaluating, benchmarking, or improving AI systems. Strong practical proficiency with Python is required, along with hands-on experience using PyTorch, JAX, TensorFlow, scikit-learn, XGBoost, LightGBM, or comparable ML tools. The AI interview includes technical and integrity checks, so expect follow-up questions about specific projects, methodological choices, failure detection, and your personal contribution.
Evidence and preferred background
Applicants are asked to submit their AI-training experience and public technical evidence. Relevant evidence may include GitHub repositories, public model implementations, Kaggle, DrivenData, or AIcrowd results, Hugging Face models or datasets, publications, technical reports, open-source contributions, benchmarks, experiments, or technical projects. Preferred experience includes creating tasks, rubrics, graders, reference answers, or benchmark problems, reviewing other trainers' submissions, working on long-horizon or agentic AI evaluations, and earning strong technical competition results.
Logistics
The role is global, fully remote, and contractor-based, with a flexible schedule and a stated commitment of 15+ hours per week. The listing does not explicitly describe the work as asynchronous. Compensation is output-based and depends on tasks meeting project specifications and quality requirements, while task availability and completion time may vary.
Pay band: $60–120/hr