Create and assess insurance challenges for AI training, using your industry knowledge to test how models handle underwriting, claims, policy interpretation, risk, and regulation. You will write realistic scenarios, review model answers against industry standards, and work with insurance and technical specialists to shape the difficulty and learning value of the content. The role is remote and asynchronous.
Mercor keeps this listing open to build a pool of machine learning practitioners for AI training and evaluation projects. You may write problems from systems you have shipped, assess model-generated code, or spot approaches that could fail in production. Apply with your resume, confirm where you work from, and complete an AI interview that takes about 20 minutes. Project invitations can arrive within a week or several months.
Review AI-generated math solutions for correct calculations, sound methods, and clear reasoning. You will compare responses, verify quantitative claims with public sources, and write explanations that show how to reach a valid result. The role calls for advanced expertise in pure and applied mathematics and strong English writing. No project is active now; qualified experts may hear about future assignments through the expert network.
Contribute intelligence expertise to an AI training and evaluation project centered on intelligence operations and national security. You will judge AI-generated scenarios and related materials for accuracy, realism, and operational relevance. The project especially seeks people with experience at civilian intelligence agencies such as the CIA, NSA, MI6/SIS, GCHQ, or equivalent organizations.
Apply machine learning to real atomistic, electronic, and structural materials data in a remote, project-based role. Work asynchronously over the next 2–3 weeks, with an expected 10–20 hours per week for projects you accept. You will build and assess models, analyze simulation outputs, and document research methods. The work also offers exposure to an early-stage, Y Combinator-backed company and material for your research portfolio.
Use your legal practice experience to help a leading AI lab assess and improve how its models handle real legal work. You will review legal tasks and model responses, write task instructions and reference solutions, and create evaluations alongside research teams. The role requires 8 to 15 years of post-qualification practice, U.S. bar admission, and a Bay Area residence for a hybrid schedule. Cincinnatus LLC employs this full-time W-2 role for an initial six-month engagement.
Create practical development exercises in Eclipse IDE, then record your workflow and explain the task in SuperAnnotate. Examples may cover Java project setup, JDK configuration, debugging, dependency management, tests, Git, or build and runtime troubleshooting. You will need a computer setup that can run Eclipse and the project materials, and the interview takes place in English.
Review marine geology and sedimentology material for technical accuracy as part of a project that trains AI systems. You’ll assess model responses, improve reference content, and document unclear cases involving marine sediments, core samples, and ocean-floor layers. The work calls for relevant domain expertise, not prior AI experience.
AfterQuery engages network engineers as contractors to create and assess scenarios used to evaluate frontier AI models. You will build examples that span network design and outage diagnosis, explain the reasoning behind sound solutions, and catch configurations that could fail in production. The work runs remotely on an async schedule, with a commitment of at least 10 hours per week.
A senior strategy specialist will assess materials on the luxury automotive sector and judge how executives at brands such as Ferrari, Porsche, Bentley, and McLaren might respond. The work calls for a C-suite perspective on market analysis, recommendations, and the realities of brand positioning, pricing, production, and client relationships. The engagement centers on written feedback, with an optional short debrief call.
Review AI-generated aviation material using expertise in commercial piloting and flight operations. You will check instrument approach procedures, cockpit briefings, FAA charts, and terminal procedures, then document corrections and feedback. Prior AI experience is not required; aviation knowledge is the focus.
Create clinical questions and case studies for AI training, then assess model answers for clinical accuracy and alignment with medical assisting standards. The work draws on hands-on experience across clinical and administrative practice, covering subjects such as patient intake, vital signs, procedures, EHR documentation, lab work, and pharmacology basics.
Assess how well a large language model handles Peninsular Spanish across translation, summaries, dialogue, and grammar correction. You will refine Spanish-language data, mark linguistic features, and share detailed findings with AI researchers to support model training and improve its Spanish comprehension and writing.
Review vehicle records and visual evidence for a project that helps train AI systems. You’ll assess damage and condition, document accident findings, and provide written and spoken feedback. The contractor role averages about 10 hours per week, with flexible days and hours, including weekends if you choose. No previous AI experience is needed.
Create and assess patent prosecution material used to evaluate AI models. You’ll develop scenarios from prior art analysis through claim drafting and office action responses, then write reference answers and judge model work for legal soundness and technical accuracy. The role is remote, asynchronous contract work with a minimum commitment of 10 hours per week and no fixed schedule.
Use your own Mac running macOS Sonoma 14.5 to work with the desktop system, including Finder, System Settings, window and Space management, and bundled apps. You must already use Sonoma 14.5 as your primary desktop operating system and have a display setup that can capture screenshots above 2.5 megapixels. The company does not supply hardware, software licenses, or virtual machines.
Contribute property claims expertise to a project that trains AI systems on real-world loss evaluation and restoration work. You’ll review digital claim materials, assess coverage and estimates, and document your reasoning. The contractor role averages about 10 hours per week, with flexible hours and days, including weekends if you choose. No AI experience is needed.
Use your property management expertise to shape AI training content. You will write and assess questions and case studies on real-world operations, from tenant relations and maintenance to lease administration and housing compliance. Your feedback will help models produce answers grounded in established practices and legal standards.
Review Neuron Kernel Interface (NKI) development work that supports training and evaluating frontier AI models. You will judge kernel correctness, CUDA-to-NKI migration fidelity, Trainium performance choices, and numerical consistency across GPU and Trainium hardware, then write clear feedback against defined rubrics. The work calls for hands-on NKI experience with AWS Trainium or Inferentia2.
A senior claims investigation expert will review evidence for signs of staged losses and insurance fraud, then help train AI systems with grounded assessments. The work draws on claims, forensic, or investigative expertise rather than prior AI experience. This global remote contract averages about 10 hours per week, with flexible days and hours, including weekends if preferred.
Location and eligibility
Remote · Eligibility not specified
Engagement type
Contract
Compensation
$20–$60/hr
Special investigations and staged-loss
fraud detection from claim evidence
AI trainingView job
Showing 20 active roles of 801. Dates reflect what the source published, not when AIWorkJobs last synced.
Sources: ten specialist AI work platforms. AIWorkJobs does not verify every claim made by a source. Report a listing that looks wrong through the contact page.