Catalog updated: . Page 2 of 6; 104 listings in this area.
Catalog order, without personalized ranking. Compare requirements with the person’s stated experience, education, skills, working languages and country. Unknown requirements are not proof of eligibility. Follow the actual page links below for more roles.
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is [**arXiv:2509.26574**](https://arxiv.org/abs/2509.26574) and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Five areas. We match narrowly: you…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is [**arXiv:2509.26574**](https://arxiv.org/abs/2509.26574) and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Four areas, three in torsionful…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is **[arXiv:2509.26574](https://arxiv.org/abs/2509.26574)** and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Seven areas. We match narrowly:…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is **[arXiv:2509.26574](https://arxiv.org/abs/2509.26574)** and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Seven areas. We match narrowly:…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is **[arXiv:2509.26574](https://arxiv.org/abs/2509.26574)** and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Twelve areas. We match narrowly:…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Doctorate / PhD
Named languages
Not listed
Listed skills
Not listed
## About the work **CritPt** is a public benchmark of research-level physics challenges, built to test whether frontier AI models can carry out genuine physics research reasoning rather than textbook problem solving. The benchmark paper is **[arXiv:2509.26574](https://arxiv.org/abs/2509.26574)** and we recommend reading it before applying. It will tell you quickly whether this work interests you. We are engaging physicists to work on research-level physics problems in their own subfield. Depending on where your publication record fits, that can mean creating problems, solving them, reviewing completed work, or auditing it. We agree the specific assignment with you once you are matched to an area. This is research-grade work rather than volume work. Whatever you produce has to be complete enough for another specialist in your subfield to follow and verify independently, so written reasoning is part of every assignment. ## Research areas in this panel Nineteen areas. We match narrowly:…
ethos · Data analysis & BI, Other data science roles, Other science roles
Pay
100 per hour (currency unconfirmed)
Country
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Confirm work and payout eligibility in the Ethos terms.
Experience
4 years minimum
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
About the role We're working with a leading foundational AI lab to find experienced data scientists and ML engineers who can help train their latest language model on data science documentation tasks. We're looking for practitioners with 4+ years of experience to review and assess documents - judging whether the content and structure of each reflect the material they would genuinely encounter in day-to-day data science practice. Compensation: $100/hour Commitment: Flexible, up to 20 hours/week Location: Fully remote Start date: ASAP Qualifications: 4+ years of experience as a data scientist or ML engineer Current, hands-on familiarity with the documents your team writes and reviews Excellent written communication and attention to detail
Remote work is listed. Required residence: Canada, United States. All listed country and residence restrictions must be satisfied. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work We are building expert-authored forecasting and research data used to evaluate and improve AI models on real political and financial analysis. Each task asks a practitioner to reason from a fixed evidence cutoff — no hindsight — and document the judgment a strong analyst would have made at that moment. This role covers political report authorship: elections, polling, and political-risk events. ## What you will do - Author point-in-time political reports and forecasts from a stated evidence cutoff - Make and document directional judgments about polling, win probabilities, and market-implied odds - Explain the causal mechanism behind an event's expected political and market impact - Review model-generated analyses and grade them against your own standard ## Who we are looking for - Local political analysts with hands-on experience on specific races in swing states: current or upcoming Senate, governor, or statewide contests, or recent cycles in the same state -…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for scientific judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for researchers who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match researchers in the network to projects like: - Writing a research problem from your own field, with the system, the data, and the assumptions, plus the reference answer a model's attempt gets graded against - Grading model-written analysis and interpretation: whether the method suits the question, whether the statistics support the claim, whether the conclusion follows from the evidence - Flagging where a model…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for mathematical judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for mathematicians who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match mathematicians in the network to projects like: - Writing a problem at research or competition level, with the statement and the intended solution, plus the reference answer a model's attempt gets graded against - Grading model-written proofs: whether each step follows, whether the edge cases are handled, whether an appeal to a known theorem actually satisfies its hypotheses - Flagging where a model reaches a…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for physics judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for physicists who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match physicists in the network to projects like: - Writing a problem from your own area, with the system, the assumptions, and the regime, plus the reference answer a model's attempt gets graded against - Grading model-written derivations and numerical work: whether the approximation is valid in the regime claimed, whether the limiting cases check out, whether the units and scaling survive scrutiny - Flagging where a model…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for materials science judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for materials scientists who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match materials scientists in the network to projects like: - Writing a materials problem from work you have done, with the system, the processing history, and the property target, plus the reference answer a model's attempt gets graded against - Grading model-written structure-property reasoning: whether the microstructure explains the property, whether the characterization supports the claim, whether…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for chemistry judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for chemists who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match chemists in the network to projects like: - Writing a synthesis or characterization problem from your own work, with the substrate, the conditions, and the analytical data, plus the reference answer a model's attempt gets graded against - Grading model-proposed routes and mechanisms: whether the disconnection is sensible, whether the selectivity claim is supported, whether the spectra match the structure assigned -…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for biotechnology judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for biotech scientists who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match biotech scientists in the network to projects like: - Writing a development problem from a program you worked on, with the modality, the process step, and the specification it had to meet, plus the reference answer a model's attempt gets graded against - Grading model-written process and assay development work: whether the scale-up assumptions hold, whether the potency assay measures what it claims,…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
The research teams behind the best-known AI models come to Mercor for biology judgment they can't generate on their own. Your standards and expertise become the rubric that frontier AI is graded against. This is a standing listing for biologists who want AI training and evaluation work, not a specific job opening. It stays open whether or not a matching project is running this week. New projects start on short notice, so instead of posting and pulling a listing each time, we keep one open per field and draw from it when one begins. **What the work looks like** We match biologists in the network to projects like: - Writing a research problem from your own work, with the system, the data, and the controls, plus the reference answer a model's attempt gets graded against - Grading model-written experimental designs and interpretations: whether the controls rule out what they claim to, whether the statistics suit the design, whether the conclusion follows from the figure - Flagging where…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work We are building expert-authored forecasting and research data used to evaluate and improve AI models on real political and financial analysis. Each task asks a practitioner to reason from a fixed evidence cutoff — no hindsight — and document the judgment a strong analyst would have made at that moment. This role covers political report authorship: elections, polling, and political-risk events. ## What you will do - Author point-in-time political reports and forecasts from a stated evidence cutoff - Make and document directional judgments about polling, win probabilities, and market-implied odds - Explain the causal mechanism behind an event's expected political and market impact - Review model-generated analyses and grade them against your own standard ## Who we are looking for - A senior elections forecaster, campaign strategist, polling or statistical-methodology expert, or political-risk analyst - A record of making live judgments about polling, win…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones. You will: - Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial. - Evaluate the responses that come back against a defined policy standard, and judge whether each was handled correctly. - Write the reference answer — what a correct response looks like, and the technical reasoning for why. This domain lives on the dual-use line more than most. The same casework knowledge that identifies a substance describes how it was produced; the same precursor list that supports an investigation is a shopping list read backwards. A model that refuses a routine identification or casework question is failing as badly as one that answers a…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones. You will: - Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial. - Evaluate the responses that come back against a defined policy standard, and judge whether each was handled correctly. - Write the reference answer — what a correct response looks like, and the technical reasoning for why. This domain lives on the dual-use line more than most. The same method that detects a hazardous compound at trace level defines the threshold someone would try to stay beneath; the same library entry that identifies an unknown tells you what a screen will and won't see. A model that refuses a routine method-validation question is failing as…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones. You will: - Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial. - Evaluate the responses that come back against a defined policy standard, and judge whether each was handled correctly. - Write the reference answer — what a correct response looks like, and the technical reasoning for why. This domain lives on the dual-use line more than most. The same route design that scales a medicine underpins a controlled compound; the same impurity profile that proves a clean process identifies how something was made. A model that refuses a routine scale-up or kinetics question is failing as badly as one that answers a question it…
Remote work is listed. Country eligibility unconfirmed; remote does not mean worldwide eligibility. Also check mercor country restrictions.
Experience
Minimum not listed
Education
Not listed
Named languages
Not listed
Listed skills
Not listed
## About the work Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones. You will: - Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial. - Evaluate the responses that come back against a defined policy standard, and judge whether each was handled correctly. - Write the reference answer — what a correct response looks like, and the technical reasoning for why. This domain lives on the dual-use line more than most. The same toxicology that sets a countermeasure dose describes an exposure someone might intend; the same decontamination data that protects responders tells you what defeats it. A model that refuses a routine countermeasure or PPE question is failing as badly as one that answers a…