LILT (Production) is seeking experienced software engineers to design, build, and validate multilingual benchmarking tasks. You will create native-language assets that test model robustness in multilingual settings, with a focus on encoding, locale conventions, and safe string handling. The role is remote and freelance. You will collaborate on high-signal tasks, develop reference implementations, and contribute to a rigorous quality control process. #J-18808-Ljbffr