Research Engineer, Code Generation & Model Evaluation

Help train next-generation AI systems by applying practical software engineering expertise to real-world code generation and model evaluation work.

This remote contract role focuses on improving how AI models learn, reason, and perform through high-quality coding tasks, evaluations, feedback, and annotations. Prior AI experience is not required.

• Analyze, debug, and resolve issues in codebases written in Python 3, Java, Rust, C++, Go, or TypeScript.
• Design and implement features and enhancements for code generation workflows.
• Refactor and optimize codebases for performance, maintainability, and adaptability.
• Develop and review real-world coding tasks to assess AI model accuracy and performance.
• Write clear, actionable feedback and annotations that support model training and evaluation.
• Work with open-source contributors and technical stakeholders to establish technical direction and produce robust results.
• Document technical decisions, best practices, and solutions to support transparency and knowledge sharing.

• Strong competitive programming and coding problem-analysis skills.
• Expertise in at least one of Python 3, Java, Rust, C++, Go, or TypeScript, along with solid knowledge of algorithms and data structures.
• Experience debugging, implementing features, refactoring codebases, and optimizing performance.
• A demonstrated record of open-source contributions or collaborative software-project work.
• Ability to evaluate complex constraints and consider multiple solution approaches.
• Excellent written and verbal communication skills, with the ability to explain technical details clearly.
• Careful attention to code validation and output consistency.
• Ability to work independently in a remote, collaborative environment and deliver well-documented work under tight deadlines.

• Remote, independent contractor engagement.
• Compensation is task-based, and completion time varies by experience and workflow.
• Minimum submission requirements apply, including a minimum number of task submissions each week.
• Roles are generally filled within 48 hours. Selected candidates should be ready to begin initial tasks within 24 to 48 hours after completing onboarding.

$50 to $100 per hour. Payment is based on tasks that meet project specifications.

Apply using an email address or an available third-party sign-in option. By continuing the application, candidates acknowledge the applicable candidate terms and privacy policy.