Expertise: stem_expert

We are seeking an experienced Civil or Structural Engineering expert to lead a pod of trainers building realistic, terminal-based technical tasks for Terminal Bench Science. You will review every task your trainers produce for engineering correctness, reproducibility, and grader reliability, and coach trainers to consistently deliver high-quality tasks. Main tasks listed: Review engineering tasks end to end, including problem statements, structural and geotechnical models, load cases, simulation data, computational environments, reference solutions, and automated tests; Validate tasks involving structural analysis, finite-element methods (FEA), earthquake and wind engineering, structural dynamics, geotechnical modeling, hydraulics, hydrology, transportation, and infrastructure systems; Ensure tasks require genuine multi-step engineering reasoning and reflect realistic civil and structural engineering workflows; Verify engineering accuracy across units, equilibrium, boundary conditions, load combinations, design codes, numerical stability, convergence, and safety factors.

We are seeking an experienced Chemical Engineering expert to lead a team of technical trainers developing realistic, terminal-based engineering tasks for a frontier AI training and evaluation initiative. In this role, you will oversee a pod of approximately 5–10 trainers, review chemical engineering tasks for technical accuracy, reproducibility, and evaluation quality, and ensure tasks reflect authentic chemical engineering workflows. Main tasks listed: Lead and mentor a pod of technical trainers, ensuring consistent quality, productivity, and timely delivery of chemical engineering tasks; Review engineering tasks end to end, including problem statements, reaction kinetics, thermodynamic and property data, process flowsheets, computational environments, reference solutions, and automated tests; Validate tasks involving reaction engineering, thermodynamics, phase equilibria, transport phenomena, separation processes, process simulation, process control, optimization, combustion, and electrochemical engineering; Ensure tasks reflect realistic chemical engineering workflows and require meaningful multi-step technical reasoning.

As a Task Executor – Aerodynamics, you will use Flow5 v7 to complete assigned aerodynamic analysis workflows involving foils, wings, aircraft, hydrofoils, and sails. You will receive the required geometry files, project files, operating conditions, and execution steps, and you will follow them exactly. Your work may include modifying geometries, configuring analyses, generating conforming meshes, running Type 1 or Type 2 polars, and exporting the requested results. Main tasks listed: Open and inspect foil, plane, mesh, and project files in Flow5; Build or modify foils, wings, planes, hydrofoils, and other assigned geometries; Configure operating conditions, polar settings, geometry parameters, and mesh requirements; Generate conforming meshes and run Type 1 or Type 2 polar analyses. This work may involve recording or sharing personal data (for example voice, video or account activity): read the consent and privacy terms on the official page before applying.

We are seeking experienced Electrical Engineers to design, solve, and review simulation-led engineering tasks used to train and evaluate advanced AI models. The role requires strong technical fundamentals, practical simulation expertise, and the ability to produce clear, reproducible engineering solutions. We are hiring for a PAA-based project, seeking candidates with strong hands-on experience in CAD modeling and simulation tools. Main tasks listed: Develop CAD models and perform simulation/analysis for aero, mech and elec components; Validate designs using industry-standard tools (e.g., CATIA, SolidWorks, ANSYS, MATLAB); Develop realistic electrical engineering problem statements, simulation models, calculations, schematics, and supporting technical assets; Define inputs, assumptions, constraints, operating conditions, output metrics, and objective acceptance criteria.

Remote contract work for hardware engineers on AI training projects: create rigorous, self-contained hardware engineering problems based on realistic scenarios and develop reference solutions with calculations, simulations or scripts (for example Python, MATLAB, Simulink, SPICE or C/C++).

Remote contract work for aerodynamics specialists on an AI training project: create, solve and validate CFD and aerodynamic tasks through command-line workflows, such as preparing domains and meshes, setting boundary conditions, running simulations, computing lift and drag and checking that results are physically valid. Python is used to automate the workflow.

Remote contract work for mechanical engineers on an AI training project: create, solve and validate engineering tasks across structural, thermal, mechanical and multiphysics domains, such as building computational models, meshes, materials and boundary conditions, running simulations and checking results against physical and design constraints. The listing describes the role as very coding-heavy.

Remote contract work for STEM experts on an AI training project: solve, review and validate challenging problems in your field, use Python for calculations or simulations, and check whether AI-generated answers are technically correct. Fields include chemistry, biology, mathematics, physics, earth and environmental science and robotics. The listing describes the role as very coding-heavy.

Remote full-time work for electrical engineers on an AI lab’s training-data team: design domain tasks and write reference solutions in circuit analysis and design, power systems and power electronics, and embedded and control systems, then evaluate model outputs with written technical feedback. Initial engagement of 3 to 4 months, 40 hours per week, with possible extension.

We’re looking for STEM specialists who live and breathe mathematics, physics, chemistry, engineering, data science, computer science, and scientific computing. You’ll challenge advanced language models on topics like thermodynamics, mechanics, calculus, statistics, materials science, electronics, algorithm design, and quantitative modeling—documenting every failure mode so we can harden model reasoning. On a typical day, you will converse with the model on real-world STEM problem-solving scenarios and theoretical technical questions, verify factual accuracy and logical soundness, capture reproducible error traces, and suggest improvements to our prompt engineering and evaluation metrics.