Senior Software Engineer

What You'll Do7

• Design realistic, challenging coding tasks for frontier AI models using repositories you know well.
• Craft problems that demand strong engineering skills, such as finding subtle bugs, building non-trivial features, or improving performance.
• Define each problem clearly and set up a reproducible environment for it.
• Write deterministic verifiers that accept any correct solution while rejecting incomplete, incorrect, or hard-coded answers.
• Provide a reference solution that shows one valid way to solve the task.
• Submit a minimum number of tasks each week as required by the project.
• Document your tasks and communicate technical details so others can follow your reasoning.

Requirements7

• Significant hands-on expertise in at least one of: Python3, Java, Rust, Go, C++, or TypeScript.
• Deep understanding of algorithms, data structures, and software engineering principles.
• Proven ability to debug complex systems, resolve bugs, and deliver effective optimizations.
• Experience with large codebase refactoring and legacy system modernization.
• Track record of delivering high-impact features from conception through delivery in cross-functional projects.
• Strong documentation and communication skills to explain technical concepts clearly.
• Interest in or curiosity about AI and its technical challenges. Previous AI experience is not required.

Who Should Apply

This role fits senior engineers who have spent years in one or more of Python3, Java, Rust, Go, C++, or TypeScript and can show deep work in algorithms, debugging, refactoring, and performance tuning. You should be comfortable designing problems from scratch, writing verifiers, and documenting your approach. It is less suitable for engineers who prefer only greenfield feature work or who need close direction on task design. Candidates often get rejected when they cannot demonstrate hands-on debugging of complex systems or when their examples lack specifics about refactoring and optimization. Another common reason for low fit scores is not showing a clear track record of delivering features end to end in cross-functional settings.

Salary Insight