Role Overview
Marvell offers a PhD research internship in storage systems for AI, engaging with open problems at the frontier of AI systems using real silicon, large-scale infrastructure, and production data. Interns contribute to work intended for publication, translate research into next-generation silicon, and collaborate with senior architects on storage and memory systems for AI/ML workloads.
\n
Responsibilities
- Explore next-gen storage and memory architectures for AI/ML workloads
- Work with NVMe and SPDK storage software stacks for high-performance data access
- Investigate memory tiering and storage disaggregation for large-scale AI systems
- Explore computational/near-storage techniques including quantization, compression, predictive caching, and data migration
- Prototype and evaluate storage architectures for performance, scalability, and efficiency
\n
Requirements
- Enrolled in a PhD, MS by Research, or MTech by Research program in Computer Science, Computer Engineering, electronics or Electrical Engineering, Robotics or a related field
- Strong fundamentals in computer architecture, storage systems, and operating systems
- Familiarity with storage technologies and stacks including NVMe, SPDK, SSDs, and memory hierarchies
- Understanding of ML data pipelines or workload characteristics
- Programming experience in C/C++ and Python
\n