LLM Engineering Expert

Role Overview

We are looking for experienced engineering professionals to create and validate challenging, simulation-based engineering design problems for advanced AI systems.

You will work across disciplines such as electrical, mechanical, aerospace, control systems, and robotics. The role involves developing complex engineering tasks, configuring open-source simulation tools, examining AI-generated solutions, identifying technical failures, and building objective evaluation methods.

What You’ll Do

• Create original engineering design problems with competing constraints, clear optimization targets, validated reference solutions, and objective graders.
• Build and test problem environments using open-source simulation tools and custom Python test benches.
• Review AI-generated code, outputs, and execution logs across repeated trials.
• Identify failure patterns such as incorrect simulator interpretation, premature convergence, and physically impossible designs.
• Refine task difficulty using model-performance results while keeping problems clear and technically complete.
• Collaborate with AI researchers, engineering experts, and project leads.
• Ensure physical plausibility, unit consistency, appropriate boundary conditions, and reliable convergence criteria.
• Produce clear technical documentation for each task and its evaluation process.

Requirements

• Master’s degree or PhD in Electrical, Mechanical, Aerospace, or a closely related engineering discipline.
• At least 10 years of hands-on engineering design experience.
• Strong Python scripting skills.
• Proficiency with at least one relevant open-source simulation tool, such as ngspice, PySpice, OpenFOAM, FEniCSx, CalculiX, python-control, CadQuery, build123d, OpenModelica, Cantera, or Gmsh.
• Experience working with modern LLMs or coding agents.
• Familiarity with AI evaluation concepts such as pass@k, failure-mode analysis, and nondeterministic behavior.
• Ability to review execution logs and identify reasoning, coding, or tool-use failures.
• Excellent attention to technical accuracy, documentation, and detail.
• Reliable computer and stable high-speed internet connection.
• Availability for 40 hours per week, including at least 4 hours of PST overlap.

Relevant Engineering Domains

• Electrical Engineering
• Mechanical Engineering
• Aerospace Engineering
• Control Systems
• Robotics and Systems Engineering

Engagement Details

• Remote contract open to candidates in Bangladesh, India, Indonesia, Egypt, Ghana, Nigeria, Turkey, Brazil, and Colombia.
• 12-week engagement.
• 40 hours per week with at least 4 hours of PST overlap.
• $500 per approved task. Each task is expected to require approximately 30 to 40 hours.
• Immediate start.