Senior GPU & AI Infrastructure Architect (Equity)

Roblox; California (MO); On-site; USD 243,000 - 295,000; Full time

Job summary

Roblox is seeking a GPU Hardware Engineer to own the full lifecycle of GPU hardware, from architectural assessment to large-scale fleet integration and performance tuning. You will ensure our GPU and AI accelerator workloads render and infer efficiently on our data-center hardware, with a focus on reliability and scalable tooling.

The role requires strong expertise in GPU architectures, PCIe/NVLink, and Linux-based validation, plus hands-on scripting in Python, Go, or C++.

Qualifications

• BA/BS in Electrical or Computer Engineering or related field.
• 5+ years hardware engineering focusing on GPU architecture, AI accelerators or HPC systems.
• Deep understanding of PCIe, NVLink, InfiniBand and liquid cooling for high-TDP hardware.
• Linux environment with hardware validation and testing experience.
• Proficient in Python, Go, or C++ for tooling.

Responsibilities

• Prototype next-generation GPU-accelerated hardware platforms and ensure integration with high-density compute nodes.
• Drive GPU integration, performance testing, and debugging in the fleet.
• Develop and execute validation and stress-testing for GPU-heavy servers.
• Lead firmware qualification (BIOS/BMC) and automate GPU health management.
• Collaborate with hardware vendors and triage component-level failures across hardware, driver and kernel layers.
• Build and maintain monitoring stacks (Grafana/Prometheus) to track GPU metrics.

Skills

GPU architecture

AI accelerators

HPC systems

Linux

Python

Go

C++

Kernel debugging

Education

BA/BS in Electrical or Computer Engineering

Tools

NVIDIA HGX/MGX platforms

PCIe fabric

NVLink

Grafana

Prometheus

Job description

Roblox is seeking a GPU Hardware Engineer to own the full lifecycle of GPU hardware, from architectural assessment to large-scale fleet integration and performance tuning. You will ensure our GPU and AI accelerator workloads render and infer efficiently on our data-center hardware, with a focus on reliability and scalable tooling.

The role requires strong expertise in GPU architectures, PCIe/NVLink, and Linux-based validation, plus hands-on scripting in Python, Go, or C++.