22 hours ago
Hyderābād, IndiaIntern
Responsibilities
- Work with CXL and UALink memory and interconnect technologies.
- Model and simulate large-scale AI training and inference workloads across memory and fabrics.
- Build QEMU-based full-system emulation platforms and discrete-event fabric models to validate software and silicon before hardware exists.
- Build and operate containerized AI workloads on Kubernetes, including device plugins, resource scheduling with DRA, operators, scaling, and reliability.
- Read research papers, contribute to CXL, UALink, Kubernetes, and Linux open standards, write research papers, and deliver industry talks.
- Use AI coding tools to investigate specifications and unfamiliar source code.
Requirements
- Bachelor’s or master’s degree in computer science, electronics, electrical engineering, or a related field, or equivalent experience.
- Fluency with AI coding tools such as Claude Code and Cursor, or the ability to learn them quickly.
- Strong fundamentals in data structures, algorithms, operating systems, and distributed systems.
- Programming experience in one or more of C++, Python, or Go.
- Familiarity with Linux, containers, and Kubernetes, or eagerness to learn them quickly.
- Curiosity, fast learning, ownership, problem-solving ability, and clear written communication.
Benefits
- Competitive compensation and great benefits.
- A collaborative, transparent, and inclusive work environment with tools and resources for professional growth and development.
About Marvell
Marvell designs and sells semiconductors for data infrastructure, including Ethernet switching and PHYs, optical and 5G connectivity, storage controllers, and custom compute silicon (ASICs/DPUs) used by cloud providers, telecom carriers, and enterprise OEMs. Its business model centers on selling chips and platform solutions and co-developing custom silicon with major customers. Founded in 1995 and headquartered in Santa Clara, it is a public company traded on NASDAQ (MRVL).