Description
This project-based opportunity focuses on creating challenging tasks and evaluation criteria for AI coding agents. The goal is to build a dataset to test and improve AI systems by simulating realistic developer environments, designing tasks, and verifying agent solutions. It's important to note that this isn't data labeling, prompt engineering, or writing code from scratch.

