This opportunity is for experienced Software Engineers interested in contributing to advanced AI research and evaluation projects. The role involves applying deep software engineering expertise to help improve the reasoning, problem-solving, and tool-usage capabilities of next-generation AI systems.
You will create Reinforcement Learning (RL) environments that evaluate AI models on real-world software engineering tasks such as bug fixing, feature implementation, code refactoring, and performance optimization. These environments will leverage Model Context Protocol (MCP) tools and require the creation of reproducible tasks, deterministic verification methods, and high-quality reference solutions.
Ideal Qualifications
- Strong proficiency in one or more of the following languages:
Python
Java
Rust
C++
Go
TypeScript
- Deep understanding of algorithms, data structures, and software performance optimization.
- Experience debugging complex software issues and delivering maintainable solutions.
- Robust background in feature development, codebase refactoring, and software architecture.
- Excellent written and verbal communication skills.
- Ability to work independently and collaborate effectively in remote environments.
Responsibilities
- Design and build Reinforcement Learning environments that test software engineering capabilities.
- Create tasks involving bug fixing, feature implementation, code refactoring, and performance optimization.
- Develop reproducible evaluation environments and deterministic verification systems.
- Produce high-quality reference solutions and technical documentation.
- Evaluate AI-generated solutions for correctness, efficiency, and software engineering best practices.
- Contribute expertise to improve AI model reasoning and tool usage capabilities.
- Collaborate with engineers and researchers to maintain high-quality evaluation standards.
Preferred Qualifications
- Experience working with large-scale distributed systems and codebases.
- Familiarity with AI, machine learning, or developer tools is a plus.
- Experience conducting code reviews and contributing to software engineering best practices.