Senior GPU Systems & Fabric Engineer

Bitdeer Technologies Group • San Jose, CA

13d • $180,000 - $320,000

About The Position

We are seeking a Senior GPU Systems & Fabric Engineer to serve as the critical bridge between our physical GPU/network infrastructure and the Kubernetes abstraction layer. You will be responsible for creating the high-performance 'hardware foundation' that makes AI-native cloud computing possible. This role requires deep expertise in Linux kernel internals, GPU architectures, and high-speed interconnects, as you will be tasked with transforming raw, bare-metal compute resources into scalable, resilient, and multi-tenant cloud primitives. You will drive the design of our fabric layer, ensuring that our AI workloads have the low-latency, high-bandwidth environment they require to perform at industry-leading speeds.

Requirements

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, or related field.
• 5+ years of systems engineering experience, with strong proficiency in Linux kernel internals, C, or Go.
• Hands-on experience with GPU architectures (NVIDIA H100/A100), CUDA runtimes, and distributed networking (RDMA, InfiniBand).
• Deep understanding of containerized environments and Kubernetes device plugin architecture.
• Proven track record of operating, debugging, and scaling bare-metal systems in large-scale production or HPC environments.
• Familiarity with infrastructure automation (e.g., Terraform, Ansible, CI/CD pipelines) for managing hardware lifecycles.
• Strong problem-solving skills, with the ability to navigate ambiguous performance challenges at the intersection of hardware and software.
• Excellent communication skills, with a collaborative approach to working across infrastructure, scheduling, and reliability teams.

Nice To Haves

• Experience working in high-velocity, high-growth engineering environments is strongly preferred

Responsibilities

• Architect and maintain integrations for NVIDIA/AMD GPU device plugins and Kubernetes Operators to expose hardware capabilities to the control plane.
• Configure and optimize high-performance host networking stacks, including RDMA, SR-IOV, RoCEv2, and InfiniBand, ensuring line-rate throughput for distributed AI training.
• Build and manage automated hardware remediation pipelines using DCGM telemetry to proactively identify, isolate, and reset degraded GPU/NIC components before they impact production jobs.
• Implement and manage sophisticated GPU slicing technologies (MIG, vGPU) to enable efficient multi-tenant inference workloads and maximize cluster utilization.
• Profile and tune kernel-level parameters, device drivers, and runtime libraries (CUDA, NCCL) to resolve bottlenecks and optimize containerized AI workloads.
• Collaborate with the Scheduling and Storage engineering teams to ensure topology-aware placement and efficient data movement across the fabric.
• Define and enforce operational standards for bare-metal provisioning, BIOS/firmware updates, and OS hardening within the containerized environment.
• Lead technical investigations into complex performance issues spanning hardware, fabric, and software, providing actionable architectural insights.
• Mentor team members and drive documentation standards for our evolving AI hardware stack.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior