3 days ago
Base Salary
$131k - $196k/yr
Responsibilities
- Own compute provisioning, cluster bring-up, OS images, ROCm and NIC driver stacks, and environment standardization for AI Solutions Lab systems.
- Build and maintain MLPerf-style harnesses, workload libraries, test plans, result pipelines, and dashboards for repeatable validation.
- Establish observability for cluster health, job execution, failures, utilization, and regression tracking.
- Deploy and manage Kubernetes and/or Slurm for validation and benchmark workloads.
- Own infrastructure incident response, root-cause analysis, and reliability improvements after cluster or job failures.
- Partner with architects, network engineers, and performance engineers to align lab environments with customer deployments.
- Improve operational efficiency through automation, infrastructure code, and standardized workflows.
- Support cross-functional validation and turn lab results into reusable internal and external assets.
Requirements
- MS in Electrical/Computer Engineering, Computer Science, or a related field; a PhD is a plus.
- 6+ years of software or systems engineering experience, including at least 4+ years of Linux/infrastructure experience.
- Strong Linux systems, data center, cluster operations, and development experience across Windows and Linux environments.
- Proficiency in Python and TCL scripting and automation, with familiarity in C/C++ API development.
- Experience with Ansible, Terraform, test harnesses, workload libraries, test plans, result pipelines, dashboards, observability, and Kubernetes and/or Slurm.
- Strong understanding of software development flows and production software testing practices.
- Customer-facing experience and strong interpersonal, written, and verbal communication skills.
- Preferred experience includes AI or HPC cluster operations, PyTorch DDP/FSDP or NeMo, Lustre, GPFS, BeeGFS, or Ceph, and model tuning or workload optimization.
- Linux, CKA, or similar Kubernetes certifications are preferred.
Benefits
- Eligible for bonus and equity plans, including a new-hire equity grant and annual equity refreshers.
- Employee stock purchase plan with a two-year lookback.
- Family support programs, mental health resources, and recognition and service awards.
- The role may require export-license review for access to export-controlled technology; interview use of AI tools is prohibited.
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).