Materials Engineer & Python Expert - Freelance AI Trainer

Overview

This project-based opportunity connects subject-matter specialists with AI evaluation work for major technology companies. The work is not a permanent role; instead, you contribute on discrete projects that focus on testing, assessing, and helping improve AI systems.

What you'll do

You will create computational materials science challenges that are difficult for a frontier AI model to solve. Each task must have a code-checkable answer and must depend on a specialized scientific tool rather than simple synthetic data manipulation.

• Select a core tool and build a problem that depends on its waveform processing, geophysical inversion, subsurface flow, or validated scientific data workflows.
• Develop a Python reference solution and provide any required input files, domain definitions, or model files.
• Set the correct numerical result and define an appropriate tolerance so the answer can be judged accurately.
• Run repeated batch tests against the model, then adjust the task difficulty until success rates fall within the target range.
• Submit the finished task for senior review in your specialty area and incorporate feedback to improve quality.

Working style and calibration

Task calibration takes persistence. You will refine scenarios, adjust inversion settings and solver thresholds, and observe how the model behaves. The goal is to achieve a pass rate of roughly 10% to 30% during testing. This process also builds deeper practical knowledge of the chosen scientific tool and how advanced models handle complex seismic, oceanographic, and subsurface-flow problems.

Requirements

• Background in materials science or a closely related discipline.
• At least 2 years of research, applied, or teaching experience.
• Solid Python skills for building reference solutions.
• Comfort with at least one scriptable package such as ObsPy, instaseis, pyrocko, MITgcm, xmitgcm, flopy/MODFLOW, or GeoPandas, or a willingness to learn one independently.
• Ability to design tasks that truly require a specialized solver.
• Strong written English at C1 level or higher.
• Applicants without prior experience in the named tools may still be considered if they can learn quickly and start contributing promptly.

How the project works

The usual flow is: apply, complete the qualification step(s), join a project, complete assigned tasks, and receive payment.

Time commitment and pay

During active phases, the expected workload is approximately 10 to 20 hours per week. This is only an estimate and depends on the project; it is not a guaranteed number of hours.

Compensation can reach the equivalent of up to $35 per hour, with final earnings depending on your level and how quickly you contribute. Pay levels may vary by project scope, complexity, and required expertise.

Application note

Submit your CV in English and include your English proficiency level.