Ph.D. Intern - Architecture, DSP & Systems Architecture

Job Title

Ph.D. Intern - Architecture, DSP & Systems Architecture

Role Summary

This internship places a Ph.D. candidate inside Marvell's architecture, DSP, and systems teams to develop and validate system-level designs, DSP algorithms, and network architectures for hyperscale AI infrastructure.

Work spans switch and interconnect architecture, high-speed link DSP, and end-to-end connectivity; projects are applied research intended for production silicon and deployment at scale. This role may require eligibility to access export-controlled technology; non-U.S. applicants may be subject to licensing review.

Experience Level

Entry-level internship for current Ph.D. candidates. No specific years-of-experience requirement stated.

Responsibilities

Typical responsibilities include:

• Architect, model, and evaluate system-level designs for switching, interconnect, or DSP applications using analytical models and simulation.
• Develop and optimize DSP algorithms for high-speed electrical or optical links (equalization, FEC, timing recovery, signal integrity).
• Design and analyze network architectures for AI fabrics, including forwarding pipelines, congestion control, and traffic management.
• Collaborate with analog, digital, and software engineering teams to validate assumptions against hardware and silicon measurements.
• Build behavioral models and simulations in MATLAB, Python, or C++ to evaluate tradeoffs across power, latency, and bandwidth.
• Present architectural proposals and algorithm results to engineering leadership and contribute to technical documentation and design reviews.

Requirements

Core qualifications and technical capabilities expected:

• Must-have: Hands-on research experience in computer architecture, digital signal processing, or networking systems.
• Must-have: Experience modeling and simulating complex systems using MATLAB, Python, or C/C++.
• Must-have: Strong analytical fundamentals (signal processing theory, queuing theory, information theory, or computer architecture).
• Must-have: Ability to collaborate across disciplines and communicate technical trade-offs to engineering teams and leadership.
• Nice-to-have: Familiarity with hardware description languages (Verilog/SystemVerilog), cycle-accurate simulation, high-speed serial link DSP (PAM4, coherent), programmable forwarding pipelines (P4/SONiC), or RDMA/UEC networking protocols.

Education Requirements

Currently enrolled in a Ph.D. program (doctoral candidate) in Electrical Engineering, Computer Engineering, Computer Science, or a related field with research focus in computer architecture, digital signal processing, communications systems, or networking.

About the Company

Company: Marvell Technology

Headquarters: Santa Clara, California, United States

Marvell’s semiconductor solutions serve as essential building blocks of the data infrastructure connecting our world, driving innovation across enterprise, cloud, AI, and carrier architectures. The company focuses on creating transformative technology that shapes the future.