Job Description

Meta Fundamental AI Research (FAIR) is looking for Research Scientist Interns to rethink the systems that power frontier AI research. AI workloads are changing faster than the infrastructure beneath them: large-scale training, inference, evaluation, and autonomous AI agents now compete for the same accelerators. Many of the resulting questions about how to share, schedule, and operate compute at scale have no established answers.

Our team works across the stack, from hardware-software co-design and performance modeling to cluster scheduling, resource management, and the infrastructure needs of agentic AI. The goal is to make AI research faster, more efficient, and more predictable at any scale.

In this role you will embed with a world-class, multi-disciplinary team of research scientists and engineers to pursue an open research question end to end- from problem framing through experimentation to publication. Interns are expected to drive their own project with mentorship, not execute a predefined task list. Our internships are twelve (12) to twenty four (24) weeks long, and we have various start dates throughout the year.

Responsibilities

Identify and formulate open research problems in how AI workloads use and share large-scale compute.

• Plan and execute original research that advances the state of the art in AI systems and infrastructure, from new abstractions and algorithms to rigorous empirical evaluation.
• Collaborate with research scientists and engineers to translate findings into published work and open artifacts.
• Communicate results clearly to both specialist and non-specialist audiences.

Qualifications

Currently has, or is in the process of obtaining, a PhD in computer science, computer engineering, electrical engineering, operations research, or a related field

• Must obtain work authorization in the country of employment at the time of hire and maintain ongoing work authorization during employment
• First-author publication record at a top-tier systems or ML venue (OSDI, SOSP, NSDI, EuroSys, ASPLOS, SC, MLSys, NeurIPS) or equivalent
• Proficient in Python and at least one systems language such as C++, Go, or Rust Experience with cluster schedulers or resource managers such as Kubernetes or Slurm
• Experience with workload characterization, performance modeling, or simulation of large-scale systems
• Experience applying ML to systems problems, or building infrastructure for LLM training, inference, or agents
• Intent to return to the degree program after the completion of the internship