Senior Software Engineer, Reinforcement Learning

You will create reproducible reinforcement learning environments that test how AI models solve complex software engineering workflows. Your work will cover DevOps, CI/CD, debugging, and command-line tools such as Git, Docker, GDB, AddressSanitizer, and FFmpeg.

You will also create realistic software engineering scenarios, produce golden reference solutions that show reliable ways to complete tasks, and assess solutions using sound judgment about algorithms, data structures, maintainability, performance, and scalability.

• Design tasks that evaluate AI models on DevOps, CI/CD, and debugging problems.
• Build or refine reproducible reinforcement learning environments and realistic software workflows.
• Use command-line development and debugging tools, including Git, Docker, GDB, AddressSanitizer, and FFmpeg.
• Create golden reference solutions for the workflows and tasks you develop.

This is a remote, part-time contractor role. Work is available in the listed countries, subject to applicable hiring eligibility.

• $80 to $120 per hour equivalent, paid per task that meets project specifications.
• Expected commitment: approximately 15 hours per week.
• Open to candidates in the listed countries.
• Proficiency in one or more of C++, Python, Java, Go, TypeScript, or Rust.
• Strong understanding of algorithms, data structures, and performance tuning.
• Ability to debug complex software issues and deliver maintainable solutions.
• Experience with feature development, codebase refactoring, and performance optimization.
• Strong written and verbal communication, close attention to detail, and effective collaboration in remote or cross-functional settings.
• Helpful background includes large-scale or distributed codebases, modern AI or machine learning systems, and rigorous code reviews or software engineering practices.
• Prior AI experience is not required when you bring strong software engineering expertise.

You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.

The hiring platform's application for this job asks for your resume, your LinkedIn profile, your phone number, your location and answers to 8 screening questions.

AI training is the human work behind systems that learn from examples, including software tasks, model responses, and evaluations. People with strong technical expertise help define reliable examples and judge whether AI systems solve difficult problems correctly.