Electrical Engineer - AI Model Evaluation

Overview

DataAnnotation is looking for an experienced electrical engineer to evaluate how frontier AI models handle real electrical and electronics engineering work. This is a contract opportunity focused on teaching AI to reason correctly about circuits, systems, and physical constraints. Your engineering judgment and written rationale will help shape the quality of future engineering tools.

Contractors can choose which projects to work on and when, and work on their own schedule from their own computer.

Compensation: Projects are paid hourly up to $40-$125 USD per hour, with bonus rates available on some projects.

Responsibilities

• Design challenging, realistic engineering problems based on your own experience, such as specifying a circuit, stabilizing a control loop, or sizing a power stage.
• Test those problems against frontier AI models and evaluate the outputs against a professional engineering standard.
• Compare two model versions on the same prompt and determine which performed better, including where each fell short.
• Review model-generated derivations, schematics, component selections, and calculations, then explain your reasoning for the evaluation.
• Identify concrete failures with evidence, such as incorrect assumptions, sign or unit errors, unphysical results, invalid topologies, fabricated datasheet values, or off-brief interpretations.
• Contribute across analog, RF, and mixed-signal circuits, electromagnetics and antennas, control and systems, devices and materials, signal processing, photonics, and power and energy systems.

Qualifications

• Fluency in English at a native or bilingual level.
• At least 2 years of industry experience in circuit or electronics engineering.
• Background in electrical engineering, electronics engineering, or a closely related discipline.
• Strong command of circuit analysis and design tradeoffs, with the ability to defend component-level and system-level decisions.
• Ability to explain why an answer is wrong, not just that it is wrong.
• Comfort using AI tools, or the ability to learn them quickly.
• A current, in-progress, or completed bachelor's degree is preferred but not required.