STEM Researcher - Computational Fields

This role is for one of our clientsCompensation: $60-$90 per hourJoin a pioneering AI initiative focused on developing the next generation of evaluation benchmarks for frontier AI models. We are seeking researchers from computational STEM disciplines—as well as computationally intensive social sciences and humanities—to bring the rigor of real-world research into AI evaluation.In this role, you will transform scientific methodologies such as experimental design, hypothesis testing, and data-driven analysis into sophisticated, multi-step benchmark tasks that challenge state-of-the-art AI systems. Working closely with AI researchers, you'll help uncover subtle reasoning errors and methodological flaws that only experienced researchers can identify.This is a fully remote, full-time engagement requiring approximately 35 hours per week.RequirementsKey ResponsibilitiesDesign complex, research-oriented benchmark tasks inspired by real-world scientific workflows, including study design, experimentation, hypothesis testing, and data analysisDevelop comprehensive reference solutions using Python, notebooks, and computational tools with the rigor expected in professional researchDefine clear evaluation standards that distinguish sound scientific reasoning from plausible but incorrect conclusionsReview AI-generated solutions, identifying methodological weaknesses, analytical errors, and flawed reasoning that experienced researchers would recognize immediatelyCollaborate with AI researchers and fellow domain experts to improve benchmark quality, consistency, and scientific rigorContribute to the continuous refinement of evaluation methodologies for advanced AI systemsRequired QualificationsMaster's degree, PhD, or equivalent practical experience in a STEM discipline, computational social science, computational humanities, or another research-intensive field involving programming and data analysisMinimum 1 year of experience in an active research role within academia, industry, government laboratories, or a similar research environmentDemonstrated experience performing computational research involving Python, data analysis, simulation, modeling, machine learning, or scientific computingStrong understanding of experimental design, hypothesis testing, statistical analysis, and rigorous interpretation of research findingsWorking knowledge of Git, integrated development environments (IDEs), and notebook platforms such as Jupyter or Google ColabExperience with AI evaluation, benchmark development, AI training, or task authoring is preferredExcellent analytical thinking, attention to detail, creativity, and the ability to solve complex, open-ended problems independentlyStrong written communication skills for documenting technical methodologies and research findingsAbility to commit approximately 35 hours per week on a consistent basisPreferred QualificationsExperience designing reproducible computational experiments or research workflowsFamiliarity with machine learning, large language models, or AI-assisted research toolsBackground in benchmark design, scientific software development, or computational research infrastructureExperience mentoring researchers, reviewing scientific work, or contributing to peer-reviewed publicationsWhy JoinHelp shape how next-generation AI systems are evaluated using rigorous scientific methodologiesCollaborate with leading AI researchers working on frontier models and advanced evaluation frameworksApply your research expertise to improve AI reasoning, reliability, and scientific accuracyContribute to impactful work that advances the quality and robustness of AI systems across multiple disciplinesEnjoy the flexibility of a fully remote engagement while working on cutting-edge AI research initiativesEqual OpportunityWe are committed to fostering an inclusive and diverse environment where all qualified applicants receive equal consideration. Reasonable accommodations are available throughout the application and engagement process.Contract & Engagement DetailsIndependent contractor engagementFully remote with flexible working hoursExpected commitment of approximately 35 hours per weekProject duration may be extended, shortened, or concluded based on project requirements and individual performanceWork does not require access to confidential or proprietary information from any current or former employerPayments are issued weekly based on approved work completedAt this time, we are unable to support H1-B or STEM OPT candidates