About the job
Meta is seeking Research Scientists to join the Evaluations team within Meta Superintelligence Labs (MSL). Evaluations are the core of AI progress at MSL, determining what capabilities get built, which features get prioritized, and how fast our models improve. As a Research Scientist, you will provide the technical capabilities to measure and understand the capabilities of our frontier AI systems. You'll work in tandem with world-class researchers to envision, develop, and validate novel evaluations that shape the future of AI capability measurement.
Responsibilities
Curate and integrate publicly available and internal benchmarks to direct the capabilities of frontier model development
Develop and implement evaluation environments, including environments for novel model capabilities and modalities
Collaborate with external data vendors to source and prepare high-quality evaluation datasets
Execute on the technical vision of research scientists designing new benchmarks and evaluations
Build robust, reusable evaluation pipelines that scale across multiple model lines and product areas
Contribute to evaluation tooling that measures the quality and reliability of evaluation suites
Qualifications
Minimum
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
PhD degree in Computer Science, Machine Learning, or a related technical field
3+ years of experience in machine learning engineering, machine learning research, or a related technical role
Proficiency in Python and experience with ML frameworks such as PyTorch
Experience identifying, designing and completing medium to large technical features independently, without guidance
Proven success in software engineering practices including version control, testing, and code review practices
Ability to work independently and adapt to rapidly changing priorities
Preferred
Publications at peer-reviewed venues (NeurIPS, ICML, ICLR, ACL, EMNLP, or similar) related to language model evaluation, benchmarking, or deep learning
Hands-on experience with language model post-training and deep learning systems, or building reinforcement learning environments
Experience implementing or developing evaluation benchmarks for large language models and multimodal models (e.g., vision-language, audio, video)
Experience working with large-scale distributed systems and data pipelines
Familiarity with language model evaluation frameworks and metrics
Track record of open-source contributions to ML evaluation tools or benchmarks