Senior Hardware Engineer - GPU & AI Infrastructure

Every day, tens of millions of people come to Roblox to explore, create, play, learn, and connect with friends in 3D immersive digital experiences– all created by our global community of developers and creators. 

At Roblox, we’re building the tools and platform that empower our community to bring any experience that they can imagine to life. Our vision is to reimagine the way people come together, from anywhere in the world, and on any device. We’re on a mission to connect a billion people with optimism and civility, and looking for amazing talent to help us get there. 

A career at Roblox means you’ll be working to shape the future of human interaction, solving unique technical challenges at scale, and helping to create safer, more civil shared experiences for everyone.

As a member of the Infrastructure Foundation Hardware Engineering team, you will play a key role in enabling our mission to 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 our GPU and AI accelerator ecosystem. You will be responsible for the full lifecycle of GPU hardware, from initial architectural evaluation and firmware qualification to large-scale fleet integration and performance tuning. You will ensure that Roblox’s massive-scale rendering and ML workloads run on the most optimized and stable hardware possible.

You Will:

• Architect & Prototype: 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.


• GPU Optimization: Drive the integration, performance testing, and debugging of GPUs in our fleet, focusing specifically on hardware-level optimizations, driver tuning, and thermal/power management.


• Validation & Certification: Develop and execute rigorous evaluation and stress-testing strategies for GPU-heavy server platforms to ensure they meet Roblox’s unique demands for real-time rendering and low-latency AI inference.


• Firmware & Systems: Lead firmware qualification (BIOS/BMC) and troubleshooting, implementing automation systems to manage GPU health, firmware updates.


• Vendor Collaboration: Provide technical guidance and deep-dive feedback to hardware vendors. Lead critical investigations into component-level failures, triaging issues across the hardware, driver, and kernel layers.


• Observability: Build and maintain advanced monitoring stacks (Grafana/Prometheus) to track GPU metrics like HBM utilization, thermal throttling events, and PCIe bandwidth saturation.

You Have:

• Education: BA/BS Degree in Electrical Engineering, Computer Engineering, or related field with equivalent practical experience.


• GPU Expertise: 5+ years of hardware engineering experience with a specific focus on GPU architecture (NVIDIA HGX/MGX platforms preferred), AI accelerators, or high-performance compute (HPC) systems.


• Deep Technical Knowledge: In-depth understanding of modern data center technologies, including PCIe fabric, NVLink, InfiniBand, and liquid cooling systems for high-TDP hardware.


• Testing Skills: Hands-on experience testing and validating CPU, Memory (HBM/DDR5), Storage (NVMe), and high-speed networking subsystems in a Linux environment.


• Programming: Proficiency in Python, Go, or C++ for developing hardware validation tools and automation scripts.


• Systemic Debugging: Expert-level skills in debugging complex server issues remotely, with the ability to analyze kernel logs, hardware registers, and bus-level captures.

You Are:

• A Problem Solver: Decisive and effective at tracking hardware issues from identification through to fleet-wide resolution.


• A Communicator: Excellent oral and written communication skills; able to translate complex hardware constraints into actionable insights for software teams.


• Collaborative: Strong interpersonal skills with the ability to lead cross-functional projects with Data Center Ops, SRE, and external vendors.


• Adaptable: Willing to travel occasionally to data centers or vendor sites to oversee hardware deployments or "first-of-a-kind" builds.