Hardware Engineering Manager — GPU Systems & HPC

Bolt Graphics; Sunnyvale (CA); On-site; USD 200,000 - 250,000; Full time

Benefits offered by this job

Medical, Dental, & Vision - 100% paid

Equity - Stock Options

401(k) match

WFH Hardware

Job summary

Bolt Graphics in Sunnyvale, CA, is seeking a Hardware Engineering Manager to own the product lifecycle and system-level deployment of the Zeus GPU platform, moving beyond legacy architectures.

You will bridge silicon, system hardware, and operations to deliver PCIe accelerator cards and 2U HPC servers, lead hardware definition, manage ODM/EMS, and drive the hardware ecosystem to displace traditional GPU clusters.

Qualifications

• 5+ years defining or engineering high-performance PCIe-based add-in cards (AIC) or complex server systems.
• Deep understanding of PCIe Gen 5.0/6.0 electrical integrity and signal routing.
• Memory architectures (DDR5, LPDDR5X, or HBM) and large memory scaling for FP64 simulations.
• Experience with embedded processors (RISC-V/ARM), Linux-based system architectures, and high-speed networking fabrics (400Gbps+).
• Experience taking hardware from tape-out to production.

Responsibilities

• Own the hardware roadmap and form-factor optimization for PCIe accelerator cards and 2U rack-mount servers.
• Collaborate with firmware and platform teams to ensure standalone Linux operation with BMC/IPMI.
• Lead cross-functional execution from tape-out through validation to mass production.
• Coordinate with ODMs and suppliers to optimize BOM and manufacturing resilience.

Skills

High-speed PCIe hardware

PCIe Gen 5/6

RISC-V familiarity

Embedded & networking fundamentals

Education

BS/MS in Electrical/Computer Engineering

MBA (valued)

Tools

RISC-V

Linux-based systems

BMC/IPMI

Job description

Bolt Graphics in Sunnyvale, CA, is seeking a Hardware Engineering Manager to own the product lifecycle and system-level deployment of the Zeus GPU platform, moving beyond legacy architectures.

You will bridge silicon, system hardware, and operations to deliver PCIe accelerator cards and 2U HPC servers, lead hardware definition, manage ODM/EMS, and drive the hardware ecosystem to displace traditional GPU clusters.