Skip to main content
Xperteez Technology logo

AI Trainer (ML Engineering) | Remote - Contract

Xperteez Technology
2 days ago
Contract
Remote friendly (International City, AE)
Worldwide

Xperteez Technology is on the hunt for an AI Trainer to help advance the next generation of AI systems. This contract role is global and fully remote, with a flexible schedule and a commitment of roughly 15 hours per week. The listing notes a fully remote setup, though the header mentions a hybrid pattern in International City, AE. If you’ve built hands-on AI-training or model-evaluation experience, this could be a good match for you.

The work centers on using your ML know‑how to design realistic, challenging tasks for advanced AI models, then assess model-produced outputs against clear standards. You’ll create task instructions, reproducible environments, reference solutions, and evaluation criteria, providing the kind of rigorous benchmarks that help teams improve AI performance. This role is especially suitable for someone who has already done substantive AI-training work and wants to apply that experience to a real project with measurable quality criteria.

Expect a setup where you balance independence with impact. You’ll craft realistic engineering tasks, establish how solutions are evaluated, and set the bar for what counts as correct. You’ll review model-generated code, experiments, analyses, and other technical deliverables, and you’ll flag issues like data leakage, improper validation, unstable training, or reproducibility problems so the team can address them early.

Shaping the challenges for advanced AI

In this role you’ll repeatedly translate complex AI objectives into concrete tasks that stress-test state‑of‑the‑art models. You’ll build environments that others can reproduce, assemble reference answers, and define clear rubrics and graders so submissions can be judged fairly. Your day-to-day includes evaluating a broad range of outputs, from code to experimental results to technical write-ups, and spotting weaknesses that could undermine results, then documenting actionable improvements.

With your prior experience, you’ll help ensure tasks are realistic, solvable, and aligned with project goals. You’ll also help keep processes transparent by outlining what success looks like and how to verify it, which in turn supports consistent quality across different contributors.

Essential qualifications

  • Demonstrated paid experience in training, evaluating, benchmarking, or improving AI systems.

  • Strong Python skills demonstrated through hands-on work.

  • Hands-on usage of ML tools such as PyTorch, JAX, TensorFlow, scikit-learn, XGBoost, or LightGBM (or similar frameworks).

  • Public proof of skill to back up your claims, such as links to code or models.

Nice-to-have background

  • Experience building AI-training tasks, reference answers, rubrics, graders, or benchmark problems.

  • Experience reviewing submissions from other AI trainers.

  • Kaggle Expert, Master, or Grandmaster status, or strong results in other technical competitions.

  • Publications or open-source contributions in machine learning.

  • Background with long-horizon or agentic AI evaluations.

  • Ready to start immediately.

Pay, benefits and practical details

The role is contract-based with remote work, and you’ll be balancing about 15 hours per week with a flexible schedule. The posting envisions a global, fully remote arrangement, while the header notes a hybrid setting in International City, AE. Compensation is output-based, meaning you’re paid for tasks that meet project specifications and quality standards, though task availability and completion times can vary.

Advice for applicants

Lead with your AI-training experience and show off your public proof of skill right away. In your resume or application, place the strongest evidence at the top, GitHub repos, public model implementations, competition results, or Hugging Face models or datasets.

Make sure your Python and ML tooling chops are easy to spot. Mention concrete projects where you used PyTorch, TensorFlow, JAX, scikit-learn, XGBoost, or LightGBM, and describe the outcomes in terms of metrics, reproducibility, or deployable results.

Prepare to discuss how you handle tricky evaluation scenarios. Be ready to talk through data leakage, validation issues, unstable training, or reproducibility problems you’ve identified and how you resolved them in past work.

If you can start right away, say so and outline what onboarding or initial availability would look like. The team values immediacy when the fit is right, so be explicit about when you could begin and what you’d need to hit the ground running.

How to apply

To apply, submit your detailed application along with evidence of your AI-training experience and public technical materials. Expect an AI interview that includes technical checks and integrity verification, followed by a hiring-manager review. If you’re selected, you’ll go through project onboarding and receive your first assigned task.