Remote Native-Language AI Benchmark Engineer

LILT is seeking experienced software engineers to design, build, and validate benchmarks for multilingual models. This remote, freelance role focuses on creating high-quality tasks in the candidate’s native language, evaluating agents, and writing deterministic verifier scripts.

You’ll work across data, prompts, and evaluation rubrics, collaborating with a global linguistics and ML team. Ideal candidates have 5+ years of software engineering experience, strong Python and shell skills, and a deep