About Marvell
Marvell’s semiconductor solutions are the essential building blocks of the data infrastructure that connects our world. Across enterprise, cloud and AI, and carrier architectures, our innovative technology is enabling new possibilities.
At Marvell, you can affect the arc of individual lives, lift the trajectory of entire industries, and fuel the transformative potential of tomorrow. For those looking to make their mark on purposeful and enduring innovation, above and beyond fleeting trends, Marvell is a place to thrive, learn, and lead.
Your Team, Your Impact
We're building an enduring AI ecosystem at the intersection of machine learning and hardware engineering — systems that learn from how engineering actually plays out in the real world, and get measurably better over time. The domain is analog/mixed-signal IC design: a sparse, deeply structured, physics-governed problem where the ultimate feedback signal is silicon itself. We're assembling an unusually broad team — ML researchers, systems engineers, mathematicians, and circuit experts — to build foundational technology meant to compound for years, not ship a one-off tool.
What You Can Expect
- Teaching machines to represent and understand structured engineering artifacts (graphs, netlists, layouts, physical models).
- Building learning and optimization systems that search vast design spaces and converge on solutions verified against real physics.
- Designing the data, modeling, and compute backbone that lets all of the above scale and stay reproducible.
- Creating fast learned models that stand in for slow, expensive physical simulation.
- Building the agentic and tooling layer that lets these systems assist — and increasingly automate — expert engineering work.
You don't need to be an expert in all of it. You need world-class depth in something, the range to learn an adjacent domain fast, and the ambition to build something lasting.
Backgrounds that aren't the "obvious" fit but map onto this problem: sparse linear algebra, model compression/quantization, graph theory and spectral methods, convex & non-convex optimization, evolutionary computation, information & coding theory, compilers and intermediate representations, distributed systems, scientific/physics-informed ML, and theoretical/mathematical work that you suspect "should" apply to structured physical domains. If you've ever looked at circuits — or any sparse, relational, physical system — and seen math others didn't, we want to talk.
What We're Looking For
· MS/PhD in CS, EE, CompE, Applied Math, Physics, or equivalent research/industry track record.
· Deep expertise in at least one of: machine learning, large-scale systems, applied mathematics/optimization, or analog/mixed-signal design.
· Strong software fundamentals; fluent in Python (PyTorch a plus); comfortable turning research into robust, tested code.
· Evidence you can go deep in an unfamiliar domain — papers, projects, open source, or shipped systems.
· Bias toward building things that endure: clean abstractions, reproducibility, scale.
Expected Base Pay Range (USD)
202,900 - 300,260, $ per annum
The successful candidate’s starting base pay will be determined based on job-related skills, experience, qualifications, work location and market conditions. The expected base pay range for this role may be modified based on market conditions.
This role is eligible to participate in Marvell's bonus and equity plans, including a new hire equity grant and annual equity refreshers, ensuring you are rewarded for your contributions and remain invested in the company's long-term success.
Additional Compensation and Benefit Elements
Marvell is committed to providing exceptional, comprehensive benefits that support our employees at every stage - from internship to retirement and through life’s most important moments. Our offerings are built around four key pillars: financial well-being, family support, mental and physical…