Remote, task-based AI safety work for nuclear experts: write single-turn prompts labelled benign, dual-use or adversarial, evaluate how AI models respond against a defined policy standard and write reference answers with the technical reasoning behind them.
Remote, task-based AI safety work for radiological safety practitioners: write single-turn prompts labelled benign, dual-use or adversarial, evaluate how AI models respond against a defined policy standard and write reference answers explaining the correct response.
Remote part-time contract work (about 10–20 hours per week) for a research project with a leading AI lab: on your own Mac, carry out and narrate real research workflows so the work can be used to benchmark and improve AI models. The work is done on your own computer: read the privacy terms on the official page before applying.
We are hiring US-based academic chemists to support the curation and evaluation of high-quality chemistry data for frontier AI models and research laboratories. This is the Senior Reviewer track of the project. You will review complex chemistry problems and AI-generated solutions, resolve disagreements among reviewers, develop evaluation frameworks, and serve as a subject-matter advisor to client teams assessing AI model outputs.
We are seeking an analytical researcher and red-teaming specialist to build evidence-based web-research benchmark tasks that reliably expose weaknesses in state-of-the-art (SOTA) AI models (e.g., Claude Opus 5 Max). In this role, you will discover hard-to-surface facts, build precise research tasks around them, test them directly against leading LLMs, and diagnose the exact failure mechanisms in the model’s search and reasoning paths. Main tasks listed: Obscure Fact Discovery: Deeply research niche topics to identify factual ground truths that cannot be surfaced via simple, direct search queries; Source Verification: Rigorously verify facts against primary or highly credible sources, cross-check for contradictory evidence, and document exact source URLs; Task Construction: Draft clear, answerable benchmark tasks designed specifically to expose structural weaknesses in model research strategies; Red-Teaming & Model Testing: Execute tasks against designated SOTA models to confirm they reliably trigger intended model failures.
We are seeking Life Sciences Subject Matter Experts to develop advanced scientific reasoning tasks for LifeSciBench. Unlike coding-focused benchmarks, this role emphasizes deep biological knowledge, scientific interpretation, and rigorous reasoning. You will create challenging questions based on realistic scientific scenarios, experimental results, biological data, figures, literature-style evidence, and research problems. This is a non-code benchmark: tasks are not executed, and no programming is required. Main tasks listed: Design novel, challenging life sciences problems requiring advanced scientific reasoning; Create tasks grounded in realistic research scenarios, experiments, biological datasets, figures, sequences, structures, or scientific evidence; Develop clear questions with well-defined and objectively defensible answers; Write rigorous expert reference answers explaining the scientific reasoning required.
We are seeking Computational Life Sciences experts with strong scientific programming skills to develop complex, realistic tasks for Agentic Life Sciences. You will design tasks where AI agents must independently navigate scientific data, write and execute code, use computational tools, troubleshoot intermediate results, and produce scientifically valid final outputs. The goal is to evaluate whether AI systems can perform authentic multi-step scientific work — not simply answer scientific questions.
We are seeking Life Science experts to contribute advanced scientific expertise to data curation and evaluation for frontier AI models and labs. This senior reviewer role focuses on reviewing complex life science problems, developing evaluation frameworks, and advising client teams assessing AI-generated scientific outputs.
We are seeking experienced Earth Sciences experts to develop realistic, terminal-based scientific and computational tasks for a frontier AI training and evaluation initiative. In this role, you will design challenging scientific tasks involving climate science, atmospheric processes, geophysics, oceanography, geology, hydrology, remote sensing, and environmental modeling. Main tasks listed: Design realistic, multi-step scientific tasks based on authentic Earth Sciences, climate, geospatial, environmental, and geophysical research workflows; Prepare and process scientific datasets involving climate, weather, atmospheric conditions, oceanography, geology, hydrology, satellite imagery, and environmental systems; Develop computational tasks involving geospatial analysis, time-series processing, numerical modeling, interpolation, forecasting, remote sensing, and environmental risk analysis; Build reproducible, terminal-based computational environments using scientific libraries, command-line tools, and appropriate dependencies. The listing says the work can involve sensitive or graphic content.
We are seeking Life Science experts to contribute advanced scientific expertise to data curation and evaluation for frontier AI models and labs. This senior reviewer role focuses on reviewing complex life science problems, developing evaluation frameworks, and advising client teams assessing AI-generated scientific outputs.