THE ROLE: We are looking for a Senior Engineer to drive validation of next-generation AI cluster solutions. In this role, you will be at the forefront of optimizing GPU cluster, working across the full system stack to ensure our solutions meet the demanding requirements of large-scale AI workloads. The primary focus of this role is on the RDMA fabric at the heart of these systems. Understanding data flows between GPUs, NICs, and the cluster network to optimize performance at scale.
The ideal candidate brings strong expertise in GPU architectures, parallel computing, and hands-on experience validating complex, high-performance systems.
This is a high-impact engineering role for someone who thrives at the intersection of hardware, networking, and systems software and who wants to shape the performance and reliability of AI infrastructure at scale.
THE PERSON: You are a seasoned engineer who thrives on hands-on problem-solving. Equally comfortable shaping long-term strategy and diving deep into complex hardware, firmware, and driver issues. You communicate clearly across teams, take full ownership of your work, and bring the drive and work ethic to see tough problems through to resolution.
Our team is built on a culture of technical innovation and continuous career development, where your impact will be felt across performance, automation, and development.
KEY RESPONSIBILITIES: - Scalability Testing: Evaluate GPU cluster scalability through rigorous testing across diverse workloads, cluster sizes, configurations, and networking technologies including RoCE.
- Benchmarking & Performance Profiling: Develop and execute comprehensive benchmarking strategies to establish performance baselines, identify bottlenecks, and generate actionable insights for improvement.
- Performance Tuning: Implement optimization strategies across protocol enhancements, load balancing, and parallel processing to drive measurable gains in RDMA throughput, latency, and collective communications.
- NIC & Cluster Optimization: Collaborate with hardware and software teams to enhance GPU cluster performance end-to-end, with a focus on the NIC-to-network data path.
- Cross-functional Collaboration: Partner closely with hardware engineers, software developers, and system architects to integrate performance improvements into cluster architecture.
- Documentation: Produce clear, detailed documentation of performance analysis, tuning methodologies, and outcomes for both internal teams and senior stakeholders.
- Continuous Learning: Stay current with advances in GPU architectures, parallel processing, and emerging networking technologies to inform ongoing improvement efforts.
PREFERRED EXPERIENCE: - Proven experience in optimizing the performance of GPU clusters.
- RDMA network configuration, troubleshooting and performance tuning.
- Strong understanding of GPU architectures, parallel computing concepts, and network protocols.
- Proficiency in scripting languages (e.g., Python, Bash) for automation and performance analysis.
- Experience with system level performance analysis tools and methodologies for GPU clusters.
- Analytical mindset with excellent problem-solving and debug skills.
- Excellent communication and collaboration skills for effective teamwork.
ACADEMIC CREDENTIALS: - Bachelors or Masters degree in electrical or computer engineering preferred
LOCATION: Austin, TX, Santa Clara CA, Seattle WA, Secaucus NJ
This role is not eligible for visa sponsorship.#LI-SC3Benefits offered are described: AMD benefits at a glance.
