South Asian Software Engineers: Coding Tasks for AI Evaluation

About the job

WHAT WE'RE RESEARCHING We're running a paid study on the effectiveness of coding evaluation harnesses used to test AI agents. Our team is building a comprehensive suite of programming environments designed to measure agent performance accurately. This work ensures that AI coding assistants are evaluated against realistic, high-quality development scenarios. HOW IT WORKS You will walk through a series of proposed coding tasks and environments during an AI-moderated session. We will ask you to review the structure, difficulty, and realism of these programming challenges. You will provide technical feedback on the evaluation harnesses and suggest improvements to the test suites. The…