Senior Software Engineer – Remote

A company specialising in AI benchmarking seeks a senior software engineer to develop reinforcement learning environments and benchmarks that evaluate agent performance on software engineering tasks. The role involves designing engineering tasks based on real work, reviewing model outputs for errors, and creating tests and success criteria. Candidates should have experience in shipping production systems and expertise in a common programming language such as Python or Java. Strong judgement about code quality and model failures, along with independent working ability, are required. There is no need for prior AI or machine learning experience, as technical skills and attention to detail are valued most. The position offers flexible remote working, with an emphasis on producing high-quality work without managerial oversight.