Join us in helping advance the next generation of AI models!
We are looking for experienced Software Engineers to contribute to a cutting-edge AI evaluation project. This project plays a key role in improving how frontier language models understand, reason about, and generate production-quality software. You'll evaluate AI-generated solutions through the lens of an experienced software engineer, helping shape the future of AI-assisted software development.
Project Duration: Approximately 1 week
Expected Timeline: July 8–14
Expected commitment: at least 10 hours
Candidates must be located in US/ Canada due to the need for real-time collaboration with the project team.
YOE
Technical Expertise
JavaScript / TypeScript: Node.js, React, TypeScript, modern web application development
Go: Backend services, distributed systems, REST/gRPC APIs, cloud-native development (Kubernetes experience is a plus)
Python: Backend development, open-source libraries, DevOps/infrastructure automation, testing frameworks
Software Engineering
Experience working with large production codebases
Ability to quickly understand unfamiliar repositories and software architectures
Experience reviewing code for readability, maintainability, and software design
Strong engineering judgment and a solid understanding of software engineering best practices
GitHub & Open Source
Strong understanding of Git and GitHub workflows (e.g., branching, Issues, Pull Requests, and Code Reviews)
Experience navigating open-source repositories and collaborating through GitHub
Familiarity with open-source development workflows is highly preferred
Familiarity with SWE-Bench or SWE-Bench Pro is a plus
Communication
Become familiar with assigned open-source codebases and their engineering practices
Review code changes, technical discussions, and related software artifacts
Evaluate AI-generated code for correctness, quality, maintainability, and engineering best practices
Provide structured evaluations with clear technical reasoning
Collaborate with fellow technical reviewers to ensure high-quality and consistent evaluations
Note: This is not a software development role. Rather than writing code, you'll leverage your software engineering expertise to evaluate AI-generated solutions and help improve the capabilities of state-of-the-art coding models.
New
New