Senior Software Engineer, AI Evaluation / Coding Agents
Review code written by AI coding agents across real repositories, judge whether the solutions are correct, robust and maintainable, identify failure modes, and design the rubrics and feedback used to improve the model. The process includes an AI interview, a short practical code-evaluation exercise and a hiring-manager interview. Independent contractor, remote, North America only, 20 to 40 hours a week for about 3 months, with at least 6 hours of Pacific Time overlap each day.
Who it suits
5+ years of hands-on software engineering
Strong proficiency in Python, TypeScript/JavaScript, Go or another major production language
Experience in substantial real-world codebases
Strong code-review skills
Clear written explanations of why code is right or wrong
Experience using modern LLMs or AI coding tools
Skills
PythonTypeScriptGoCode reviewLLM evaluationCI/CD
Estimate your earnings on this role
A hypothetical calculation from the numbers you enter. It is not a promise of work or earnings. Most of the difference between a headline rate and what you take home is how much of your time has paid work in it.
$0 after tax and fees, expected case
Slow
Expected
Fast
Paid hours
Before tax
After tax and fees
Per month, after tax
Slow assumes 60% of your expected paid hours, fast assumes 125%. The rate used for tasks is the fee, less unpaid tasks, divided by the time one task takes.