Remote AI training and data labeling jobs

Turn a simulation, numerical solver, data-reduction pipeline or model fit from your own physics or astrophysics research into a paid AI evaluation task. $200 per approved task (about 2 hours), no cap.

Turn analysis pipelines, dynamical models or simulations from your own biology research into paid AI evaluation tasks with a Python reference solution and tests. $200 per approved task (about 2 hours), no cap. Students, postdocs and recent graduates.

Turn an electronic-structure or molecular-dynamics workflow, kinetics model, spectroscopy fit or thermodynamics calculation from your own research into a paid AI evaluation task. $200 per approved task (about 2 hours), no cap.

Student and early-career engineering researchers turn their own computational work into evaluation tasks for AI models, with a Python reference solution and tests. $200 per approved task (about 2 hours each), no cap. Not open to F-1 visa holders.

Math researchers, including research-track students, turn their own computational work into evaluation tasks for AI models, each with a Python reference solution and tests. $200 per approved task (about 2 hours, roughly $100/hour), no cap. No visa sponsorship; F-1 students not eligible.

Turn a phase-field or CALPHAD model, atomistic simulation, microstructure or diffraction analysis, or transport calculation from your own research into a paid AI evaluation task. $200 per approved task (about 2 hours), no cap.

Turn a geophysical inversion, ocean or atmosphere model, data-assimilation step or remote-sensing analysis from your own research into a paid AI evaluation task. $200 per approved task (about 2 hours), no cap.

AfterQuery contracts biostatisticians, pharmacometricians and statistical geneticists with 3+ years full time to write realistic problems and reference answers that AI labs use to test model judgment. $115/hr, paid weekly via Stripe, remote, 10+ hours a week.

PhD students, postdocs, research scientists and research-active Master's students author AI evaluation tasks from their own simulation, modelling or data-reduction work. $200 per approved task (about 2 hours), no cap.

PhD-level and PhD-pursuing researchers across the sciences, maths and engineering author AI evaluation tasks from their own computational work. $300 per approved task (about 4 hours), no cap; board headline $75/hr.

AfterQuery hires environmental scientists and engineers with 2+ years of professional experience to review, create and rank technical content for AI models. Paid hourly, fully remote and async; the board shows $100/hour.

Biochemists and molecular biologists write and grade mechanism-level life science scenarios used to test frontier AI models. Remote, async, at least 10 hours a week, $60 to $120/hour set by experience (board headline $90/hour), paid weekly via Stripe.

Mathematics teachers write instruction scenarios and worked solutions and grade AI output for accuracy and pedagogy. Remote contract work at $45 to $80/hr (board headline $63/hr), paid weekly via Stripe, 10 hour weekly minimum.

Science teachers write instruction scenarios with labs and misconceptions and grade AI output for accuracy and lab safety. Remote contract work at $45 to $75/hr (board headline $60/hr), paid weekly via Stripe, 10 hour weekly minimum.

AfterQuery hires chemical engineers with 2+ years of process, plant or product experience to author and grade process design and safety scenarios for AI evaluation. Remote, async, 10 hours a week minimum, $100–130/hour paid weekly via Stripe.

Licensed US veterinarians author and grade the clinical scenarios used to test AI veterinary reasoning. Remote, async contract work with a stated $95 to $145/hr range (board headline $120/hr), paid weekly, 10 hours a week minimum.

Geologists and earth scientists author and grade geoscience and climate scenarios that frontier AI labs use to test model judgment. Board shows $85/hr; the ad quotes $60 to $110/hr, weekly via Stripe, remote and async, 10+ hours a week.

Biologists of any specialty author and grade life science scenarios used to test frontier AI models. Remote, async, at least 10 hours a week, $60 to $120/hour set by experience (board headline $90/hour), paid weekly via Stripe.

Optics and photonics professionals with 1+ years of simulation experience design technical problems, write reference solutions and evaluate submitted work for AI training. Remote, async, project-based, $100/hour with milestone bonuses.

Senior PhD scientists (10+ years) in math, physics, chemistry or biology define and grade expert-level problems to evaluate frontier AI reasoning. Board shows $135/hr; the ad quotes $100 to $170/hr by experience. Remote, async.

A one-week, project-based AI safety evaluation for PhD or Masters-level CRISPR researchers: grade and annotate model answers on gene editing for accuracy and dual-use risk. Board shows $100/hr; the ad quotes a flat $1,000 to $1,200 for the week.

Project-based remote work for ML professionals with bioinformatics depth: models and pipelines on genomic, transcriptomic or multi-omics data. Ad states $150–200/hr, 10–20 hours a week over 2–3 weeks; the board headlines $175/hr. Needs a first-author publication.

Physics PhDs and Masters-level researchers write and check reasoning problems and review model solutions for AfterQuery. Board shows $65/hr; remote and asynchronous. Applications ask for institution, degree and a full publication list.

Chemists with a Master's or PhD write graduate-level chemistry problems and solutions and give detailed feedback on AI model answers. Remote, project-based, flexible hours, $75/hour. The form asks for affiliation, publications and degree.