Electrical Engineer, RF/Circuit Design for AI Benchmarking

Design graduate-level computational engineering challenges that evaluate how advanced AI systems use real scientific software for research-focused work. You will create and refine original tasks involving simulations, result interpretation, experiment design, and extracting hidden information from data. Role Overview

This role centers on building rigorous benchmark problems, not routine data labeling. Problems may require an AI system to complete a fully specified, multi-step workflow and produce an exact answer, or to strategically plan queries and experiments, interpret partial results, and efficiently narrow possible solutions. You will test each task with advanced AI models and refine it to reach the intended level of difficulty.

• Create challenging computational problems based on real electrical engineering, RF, microwave, and circuit-design workflows.
• Develop tasks that require simulations, data interpretation, experimental design, and inference from measurement results.
• Design problems using scikit-rf for RF and microwave network analysis, S-parameter characterization, and transmission-line modeling, and/or ngspice for circuit simulation, operating-point analysis, and frequency-response characterization.
• Create problems involving recovery of circuit parameters from measurement data.
• Write Python-based problem setups, oracle functions, and solution validators.
• Evaluate tasks against advanced AI models and revise them based on results and feedback.

• Graduate-level training in a relevant STEM discipline, including an MS, PhD, or equivalent research experience. Advanced degrees are preferred.
• Hands-on experience using at least one listed scientific software library for real research work, demonstrated through publications, open-source contributions, or professional experience.
• Strong Python programming skills.
• Deep expertise in electrical engineering and RF/circuit design, including an understanding of software limitations, edge cases, and what makes a problem challenging through reasoning rather than raw computation.
• Ability to work independently and improve problem designs in response to feedback.
• Comfort working in a Linux terminal environment and remote compute sandboxes.

• Experience with multiple relevant tools or related technical domains. Experience with other specialized electrical engineering software will also be considered.
• Familiarity with benchmark or evaluation design.
• Experience teaching scientific subjects or designing exams and problem sets.
• Experience with computational reproducibility and containerized environments.

• Remote, hourly role.
• Availability of at least 15 to 20 hours per week is required.

$70 to $85 per hour.