Chemistry Reasoning Evaluator (BS/MS/PhD Preferred)

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect people who want flexible, remote work with projects that train state-of-the-art AI systems.

This role is part of OpenTrain's mission to help contributors start and grow careers teaching AI — work that directly shapes how models reason and perform in scientific domains.

Why AI Training Work Matters

Human evaluators and annotators provide the examples and judgments that modern AI models learn from. Chemistry-focused evaluation helps models reason correctly about reactions, calculations, spectroscopy, safety, and more.

These projects are flexible, usually remote, and a fast-growing way to apply scientific training outside of traditional lab or academic roles.

The Role

You will review AI-generated chemistry responses and judge their correctness, reasoning depth, clarity, and safety. This is a contractor, part-time position paid hourly at USD 80/hour.

Work is remote and worldwide. Time commitment is less than 20 hours per week, with a minimum availability expectation of 17–20 hours/week and a preferred cadence of about 8 hours/day during active sprints.

• Work type: CONTRACTOR, PART_TIME
• Pay: USD 80 per hour
• Labeling focus: TEXT; label type: EVALUATION_RATING

What You'll Do

Use detailed rubrics to evaluate and compare multiple AI responses for scientific accuracy, reasoning quality, and communicative clarity. Draft exemplar explanations and step-by-step model solutions where appropriate.

• Assess correctness of calculations, units, stoichiometry, thermodynamics/kinetics, mechanisms, spectroscopy, and analytical methods.
• Spot subtle conceptual, methodological, and computational errors and note sources of uncertainty or approximation.
• Fact-check chemical claims using reputable public sources and provide precise, consistent referencing when required.
• Rate and compare model outputs using detailed rubrics and write clear feedback that can guide model improvement.

Requirements (Must Have)

Candidates must meet the educational and technical requirements below; we apply rubrics consistently and expect meticulous, reproducible evaluations.

• BS, MS, or PhD in Chemistry or a closely related chemical science (Top‑100 university preferred).
• Strong foundation across general, organic, inorganic, physical, and analytical chemistry, plus good lab and safety literacy.
• Ability to write rigorous, step-by-step solutions in clear C1+ English with correct notation and unit usage.
• Quantitative rigor: dimensional analysis, unit consistency, reasonable approximations, and awareness of uncertainty.
• Proven ability to apply evaluation rubrics consistently and attention to reproducibility and detail.
• Availability: minimum 17–20 hrs/week (less than 20 hrs/week total); preferred cadence ~8 hrs/day during active sprints.

Preferred Qualifications & Bonuses

These experience items are not required but make an application stronger and may be useful on complex evaluation tasks.

• Research experience, analytical writing, or debate-style argumentation background.
• Programming literacy such as Python or MATLAB and familiarity with LaTeX for clear technical notation.
• Prior data labeling, RLHF, or AI model evaluation experience is a bonus.

Compensation, Onboarding, and Logistics

This role pays USD 80 per hour as a contractor. Onboarding includes a paid 1–2 hour qualification exam and a paid 1–2 hour project exam to confirm fit and calibration.

Labeling software for the project is listed as OTHER; the task focuses on TEXT data and evaluation ratings. The position is open worldwide and remote.

• Paid qualification: 1–2 hour exam (paid)
• Paid project exam: 1–2 hour calibration (paid)
• Worldwide remote work; contractor status

How To Apply

Create a free OpenTrain account, complete your profile, and apply to this Chemistry Reasoning Evaluator role. Your application should highlight your degree, relevant coursework or research, scientific writing samples, and availability.

Successful applicants will be invited to complete the paid qualification and project exams as the next step in onboarding.