About the Company
We are a fast-growing AI research and data innovation company specializing in high-quality datasets that power advanced machine learning systems. Our mission is to accelerate global AI development through precise research evaluation, domain-expert annotations, and rigorous quality standards.
About the Role
As a Coding Agent, you will work on real-world codebases to evaluate patches, debug issues, write tests, review PRs, and validate multi-language coding workflows.
You will use tools like Cursor AI, Claude Code, GitHub Copilot, and other modern coding agents to perform high-quality engineering tasks and ensure code correctness.
Responsibilities
- Read and analyze large code repositories across multiple languages
- Validate code patches, run tests, and debug failures
- Review PRs for correctness, maintainability, and performance
- Generate tests (unit, integration, regression) across diverse projects
- Use AI-powered coding tools (Cursor, Claude Code, Copilot) to enhance workflows
- Identify design issues, edge-case bugs, and logical inconsistencies
- Produce clear annotations, reasoning steps, and code-quality evaluations
Skills & Experience Required
- Hands-on experience in one or more of the following:
- Python, JavaScript/TypeScript, Java, Go, C++, Rust, Ruby, PHP
- 1–6 years software development experience (depending on language)
- Strong debugging, testing, and PR review skills
- Experience using Cursor, Claude Code, Copilot, or similar AI coding tools
- Solid understanding of version control (Git), code reviews, and CI workflows
- Ability to read large-scale repos and evaluate patch correctness.
- Full-Stack / Backend Engineer comfortable with multi-language debugging, test creation, PR review, repo navigation, and AI-assisted coding workflows.
