GPU Hardware Engineer, Infrastructure | Roblox San Mateo

The Briefed; San Mateo (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Equity compensation for all full-time

Benefits for full-time employees

Job summary

Roblox is seeking a technical lead for GPU hardware within the Infrastructure Foundation Hardware Engineering team in California. You will own the full lifecycle of GPU hardware from architecture through firmware qualification to fleet integration and performance tuning.

You will collaborate across teams to optimize real-time rendering and AI workloads at scale. The role requires 5+ years in GPU architecture or HPC, strong Linux/driver debugging skills, and travel as needed.

Qualifications

• BA/BS in Electrical Engineering, Computer Engineering, or related field or equivalent practical experience.
• 5+ years hardware engineering experience focused on GPU architecture or HPC systems.
• Understanding of PCIe fabric, NVLink, InfiniBand, and liquid cooling for high-TDP hardware.
• Testing and validating CPU, memory (HBM/DDR5), storage (NVMe), and networking subsystems in Linux.
• Proficiency in Python, Go, or C++ for automation and tooling.
• Ability to debug server issues remotely, including kernel logs and hardware registers.
• Willingness to travel occasionally to data centers or vendor sites.

Responsibilities

• Prototype GPU-accelerated hardware platforms and ensure integration with high-density compute nodes, interconnects, and firmware.
• Drive integration, performance testing, and debugging of GPUs in the fleet with hardware-level optimizations.
• Develop evaluation and stress-testing strategies for GPU-heavy server platforms for real-time rendering and AI inference.
• Lead firmware qualification (BIOS/BMC) and troubleshooting, and build automation for GPU health and updates.
• Provide guidance to hardware vendors and lead investigations into component-level failures across hardware, driver, and kernel layers.
• Build and maintain Grafana/Prometheus monitoring for GPU metrics like HBM, throttling, and PCIe bandwidth.

Skills

GPU architecture

AI accelerators

Kernel debugging

System performance tuning

Hardware validation

Education

Bachelor's in Electrical/Computer Engineering

Equivalent practical experience

Tools

Grafana

Prometheus

Linux tooling

Job description

As a member of the Infrastructure Foundation Hardware Engineering team, you will help deliver a reliable, high-performing, and cost-efficient infrastructure that powers the world’s play. In this specialized role, you will be the technical lead for Roblox’s GPU and AI accelerator ecosystem. You will own the full lifecycle of GPU hardware, from architectural evaluation and firmware qualification to large-scale fleet integration and performance tuning. Your work will make sure Roblox’s massive-scale rendering and ML workloads run on the most optimized and stable hardware possible.

Key Responsibilities

• Prototype next-generation GPU-accelerated hardware platforms, ensuring seamless integration between high-density compute nodes, high-speed interconnects (NVLink, PCIe Gen5/6), and system firmware
• Drive the integration, performance testing, and debugging of GPUs in the fleet, focusing on hardware-level optimizations, driver tuning, and thermal and power management
• Develop and run rigorous evaluation and stress-testing strategies for GPU-heavy server platforms to meet Roblox’s demands for real-time rendering and low-latency AI inference
• Lead firmware qualification (BIOS/BMC) and troubleshooting, and build automation to manage GPU health and firmware updates
• Give technical guidance and detailed feedback to hardware vendors, and lead investigations into component-level failures across the hardware, driver, and kernel layers
• Build and maintain advanced monitoring stacks (Grafana, Prometheus) to track GPU metrics such as HBM utilization, thermal throttling events, and PCIe bandwidth saturation

Requirements

• BA/BS in Electrical Engineering, Computer Engineering, or a related field, or equivalent practical experience
• 5+ years of hardware engineering experience focused on GPU architecture (NVIDIA HGX/MGX platforms preferred), AI accelerators, or high-performance compute (HPC) systems
• In-depth understanding of modern data center technologies, including PCIe fabric, NVLink, InfiniBand, and liquid cooling for high-TDP hardware
• Hands-on experience testing and validating CPU, memory (HBM/DDR5), storage (NVMe), and high-speed networking subsystems in a Linux environment
• Proficiency in Python, Go, or C++ for building hardware validation tools and automation scripts
• Expert-level skill in debugging complex server issues remotely, including analyzing kernel logs, hardware registers, and bus-level captures
• Willingness to travel occasionally to data centers or vendor sites to oversee hardware deployments or first-of-a-kind builds

Skills

Technical Skills

• GPU architecture and AI accelerators (NVIDIA HGX/MGX)
• PCIe, NVLink, InfiniBand, and liquid cooling
• Firmware qualification (BIOS/BMC)
• Python, Go, or C++
• Linux hardware validation and kernel debugging
• Grafana and Prometheus monitoring

Soft Skills

• Decisive problem solving from identification to fleet-wide resolution
• Clear oral and written communication, translating hardware constraints for software teams
• Leading cross-functional projects with Data Center Ops, SRE, and external vendors
• Adaptability

Benefits

• Equity compensation for all full-time employees
• Benefits for full-time employees, with details on Roblox’s careers page