Research Scientist • Intern

Who you are

You are a PhD candidate with a strong background in machine learning and a passion for optimizing performance in deep learning frameworks. Your experience with PyTorch is extensive, and you understand the intricacies of compiler technologies. You have a solid grasp of distributed systems and performance optimization techniques, which allows you to contribute effectively to team projects. You thrive in collaborative environments and are eager to learn from experienced researchers while also sharing your insights. You are detail-oriented and enjoy tackling complex problems that require innovative solutions.

What you'll do

As a Research Scientist Intern at Meta, you will work closely with a team dedicated to making PyTorch faster and more resource-efficient. Your primary focus will be on advancing PyTorch 2.0 technologies, specifically enhancing the torch.compile() functionality. You will contribute to improving PyTorch's performance on GPUs, CPUs, and other accelerators, ensuring that users can leverage the full potential of the framework. You will also engage in vertical performance optimization for various models used in training and inference, collaborating with other researchers to implement cutting-edge solutions. Throughout your internship, you will have the opportunity to present your findings and contribute to the ongoing development of PyTorch, making a significant impact on the community.

What we offer

Meta provides a dynamic internship experience that lasts between twelve to sixteen weeks, with various start dates throughout the year. You will have the chance to work in a supportive environment that fosters innovation and collaboration. Interns at Meta are encouraged to explore their interests and develop their skills while contributing to meaningful projects that impact millions of users worldwide. Additionally, you will gain exposure to industry-leading technologies and methodologies, preparing you for a successful career in research and development.