Meta is seeking a Visiting Researcher through a short-term contract to join our Meta PyTorch Distributed Team. Our team's mission is to make PyTorch faster and easier to use in order to create and maintain a state-of-the-art machine learning framework that is used across Facebook and the entire industry. The key challenges in the team are extracting structure from PyTorch programs (via TorchScript, torchfx, and Lazy Tensors), generating fast CPU/GPU code (via runtimes, graph optimizations, and code generation), and enabling deployment of PyTorch (via hardware accelerators and embedding PyTorch inside other systems).This position is short-term employment for PhD students during the academic school year who will complete a minimum 6-month term.
Visiting Researcher, PyTorch Distributed Responsibilities:
- Apply relevant AI and machine learning techniques to advance the state-of-the-art in machine learning frameworks.
- Collaborate with users of PyTorch to enable new use cases for the framework both inside and outside Meta.
- Develop novel, accurate AI algorithms and advanced systems for large scale distributed training and inference.
- Currently has, or is in the process of obtaining, a Ph.D. degree in Computer Science or related quantitative field.
- 4 + years of specialization experience in one or more of the following machine learning/deep learning domains.
- Large scale training and inference ML Systems Research.
- ML theory: Basic knowledge about ML models in different modalities like LLM (Large Language Models), Vision (VITS, MVITS) and Multimodal and how scale impacts performance.
- ML systems: AI infrastructure, machine learning accelerators, high performance computing, machine learning compilers, GPU architecture, machine learning frameworks, on-device optimization.
- Must obtain work authorization in the country of employment at the time of hire, and maintain ongoing work authorization during employment.
- Experience with ML compiler development such as XLA, TVM etc.
- Experience training models at scale using PyTorch/TensorFlow/JAX.
- Publications in top tier ML or System Conferences such as ASPLOS, ICML, ICLR, KDD, NIPS, MLSys.
Facebook is proud to be an Equal Opportunity and Affirmative Action employer. We do not discriminate based upon race, religion, color, national origin, sex (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender, gender identity, gender expression, transgender status, sexual stereotypes, age, status as a protected veteran, status as an individual with a disability, or other applicable legally protected characteristics. We also consider qualified applicants with criminal histories, consistent with applicable federal, state and local law.Facebook is committed to providing reasonable accommodations for candidates with disabilities in our recruiting process. If you need any assistance or accommodations due to a disability, please let us know at firstname.lastname@example.org.