About the role#
The Compute-US team focuses on research and development for our compute infrastructure platform. This division builds large-scale cloud infrastructure that supports both public cloud products and internal systems. As a PhD intern, you will contribute to our products and research while working alongside industry experts.
What you'll need#
- Current enrollment in a PhD program for Computer Science, Computer Engineering, Software Development, or a related technical field.
- Experience in at least one of the following areas: LLM training and inference, or Physical AI infrastructure, including training, inference, simulation, RL, or synthetic data.
- Ability to work independently and complete projects from start to finish in a timely manner.
- Good communication and teamwork skills to explain technical concepts to teammates.
- Experience contributing to open-source communities like PyTorch, vLLM, or SGLang is preferred.
- Prior publications in computer science conferences such as OSDI or MLSys are preferred.
What you'll do#
- Explore and develop key technologies for the AI infrastructure stack, including AI training and inference platforms, and AI agent infrastructure.
- Build prototypes to verify the functionality and performance of new components or innovations.
- Showcase research findings and products through academic publications.
Location & details#
- Location: Seattle, Washington.
- Type: Full-time internship.
- Modality: On-site.
- Compensation: $57 per hour.
- Process: Applications are reviewed on a rolling basis.


