Remote Senior Code & Firmware Reviewer for AI Evaluation

AuraOne; United States; À distance; USD 138 000 - 165 000; Plein temps

Résumé du poste

AuraOne is seeking a remote, US-eligible contractor to review AI-generated code, reproduce results, and write minimal failing tests. You will assess outputs, explain fixes, and help target gaps for production-grade correctness.

Ideal candidates have strong software engineering experience, are adept at unit testing with common frameworks, and can communicate clear rationale. Availability of 10 hours per week is expected.

Qualifications

• Strong day-job engineering experience reading, running, and debugging unfamiliar code.
• Comfort writing concise unit tests that capture a single failure mode.
• Familiarity with a testing framework (pytest, Jest, JUnit, etc.).
• Clear written reasoning — your review note must convince another senior engineer.
• Reliable async availability for at least 10 hours per week.
• Prior code-review or technical-interview-grading experience is a plus.

Responsabilités

• Run and reproduce candidate code outputs in a sandboxed environment.
• Grade embedded systems solutions for correctness, style, and edge-case handling.
• Write minimal failing tests that demonstrate the bug a model output missed.
• Compare paired solutions and rank them with a written rationale tied to the rubric.
• Tag failure modes with severity scores.
• Document recurring code-generation failures for model improvement.
• Calibrate against gold-standard reviews to maintain inter-rater agreement.

Connaissances

Code review

Debugging

Unit testing

Software engineering judgment

Embedded systems

Outils

pytest

Jest

JUnit

Description du poste

AuraOne is seeking a remote, US-eligible contractor to review AI-generated code, reproduce results, and write minimal failing tests. You will assess outputs, explain fixes, and help target gaps for production-grade correctness.

Ideal candidates have strong software engineering experience, are adept at unit testing with common frameworks, and can communicate clear rationale. Availability of 10 hours per week is expected.