Research Scientist, Machine Learning

Meta • Paris, Ile-de-France, France • 4d ago

Meta is seeking a Research Scientist to join its AI research organization, focused on advancing core machine learning methods that power Meta's products and platforms at scale. In this role, you will conduct applied and foundational ML research with an emphasis on Reinforcement Learning and Large Language Models (LLMs), contributing to systems that serve billions of users. You will collaborate with a multidisciplinary team of researchers and engineers to translate research insights into measurable product impact.

Responsibilities

Design and execute machine learning research experiments, including model architecture design, training recipe development, and rigorous evaluation across benchmarks and production metrics
Develop and optimize ML models and algorithms targeting core research problems such as representation learning, generative modeling, or large-scale optimization
Implement clean, reusable, and well-tested research code using modern ML frameworks such as PyTorch, contributing to shared research infrastructure
Analyze experimental results to form data-driven conclusions, identify failure modes, and iterate on hypotheses in collaboration with research and engineering partners
Contribute to the preparation and submission of research findings to peer-reviewed venues such as NeurIPS, ICML, ICLR, or CVPR
Partner with cross-functional teams including product engineering, data science, and infrastructure to translate research advances into production-ready solutions
Participate in code review processes to maintain research code quality, catching subtle issues including those arising from AI-assisted code generation
Instrument experiments with appropriate logging, monitoring, and evaluation pipelines to ensure reproducibility and reliability of research outcomes
Contribute to the research community through open-source releases, technical documentation, and knowledge sharing within the team

Minimum Qualifications

Currently has, or is in the process of obtaining, a PhD in Computer Science, Statistics, Mathematics, Electrical Engineering, or a related quantitative field
2+ years of experience designing, training, and evaluating machine learning models using deep learning frameworks such as PyTorch or JAX
Experience with Reinforcement Learning and/or Large Language Models (LLMs), including areas such as RLHF, policy optimization, or LLM fine-tuning
Experience implementing and iterating on ML experiments end-to-end, from data pipeline construction through model evaluation and analysis
Must obtain work authorization in the country of employment at the time of hire, and maintain ongoing work authorization during employment

Preferred Qualifications

Demonstrated ability to contribute to open-source ML research codebases or reproducible research artifacts
Experience applying ML research to large-scale production systems, including familiarity with A/B experimentation frameworks and staged rollout practices
Experience training or fine-tuning large-scale foundation models, including familiarity with distributed training techniques such as model parallelism or mixed-precision training
First-author or significant contributions to publications at peer-reviewed ML conferences such as NeurIPS, ICML, ICLR, CVPR, or EMNLP
Interest or expertise in formal mathematics and theorem proving tools such as Lean