We are seeking a Senior Principal Network Development Engineer (IC5) to lead backend NIC qualification and New Product Introduction (NPI) for next-generation networking platforms supporting GPU- and accelerator-based clusters. This role is critical to enabling OCI’s AI superclusters by ensuring NIC technologies meet stringent requirements for RDMA performance, cluster scale, and workload isolation.

As an IC5, you will operate as a technical leader owning complex, ambiguous problem spaces end-to-end. You will drive cross-organizational initiatives spanning hardware vendors, OCI compute and networking, and data center operations, while influencing architecture and long-term roadmap decisions.

Required Qualifications

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, or related field
• 8–12+ years of experience in networking, systems engineering, or hardware validation in large-scale distributed environments
• Deep expertise in NIC architecture and advanced features (RDMA/RoCE, congestion control, SR-IOV, queueing, offloads)
• Strong understanding of distributed systems networking for AI/ML workloads (e.g., collective communication patterns, east-west traffic scaling)
• Advanced knowledge of Linux networking stack and kernel-level debugging
• Proven experience leading hardware qualification and/or NPI efforts in data center or cloud environments
• Strong debugging skills across hardware, firmware, driver, and system layers
• Proficiency in automation and tooling (Python, Bash, or similar)
Experience with performance benchmarking and traffic analysis in high-scale environments

Preferred Qualifications

• Experience with AI/HPC networking (e.g., RoCEv2, InfiniBand concepts, GPU cluster networking)
• Familiarity with distributed training frameworks (e.g., NCCL) and their network behavior
• Knowledge of PCIe, NUMA, and GPU/accelerator interconnect considerations
• Experience in hyperscale cloud environments
• Exposure to SmartNICs, DPUs, or offload-driven architectures
• Experience building validation pipelines integrated with CI/CD systems

• Background in large-scale cluster bring-up and production operations
  

IC5 (Principal) Expectations

• Technical Leadership at Scale: Owns and drives critical, cross-organizational initiatives that impact OCI AI infrastructure at fleet level
• End-to-End Ownership: Operates with high autonomy in ambiguous spaces, defining problems, strategy, and execution without requiring direction
• Cross-Org Influence: Partners effectively across compute, networking, hardware, and operations teams; influences external vendors and internal architecture decisions
• Depth + Breadth: Brings deep domain expertise in networking/NICs while understanding adjacent systems (compute, storage, AI workloads)
• Operational Excellence: Drives high standards for reliability, scalability, and performance in production environments
• Data-Driven Decision Making: Uses metrics, benchmarking, and production telemetry to guide decisions and prioritize work
• Mentorship & Multiplication: Elevates team capability through mentorship, technical guidance, and raising engineering standards
• Customer Impact Focus: Aligns engineering decisions with the needs of OCI AI customers and large-scale training workloads

• Own the end-to-end qualification strategy and execution for backend NICs supporting OCI AI clusters (RDMA/RoCE-based fabrics)
• Lead NIC NPI for AI infrastructure, from early silicon bring-up through fleet-wide deployment across OCI regions
• Define validation methodologies for high-performance, low-latency distributed training workloads (e.g., GPU collectives, east-west traffic patterns)
• Drive deep performance characterization and tuning of NICs in AI cluster environments (latency, throughput, tail latency, congestion behavior)
• Partner with NIC and silicon vendors (e.g., NVIDIA/Mellanox, Broadcom, Intel) to resolve complex hardware/firmware issues and influence feature design
• Collaborate with OCI AI/ML platform, cluster networking, and host software teams to ensure optimal integration with drivers, kernel, and user-space stacks
• Build and scale automated validation frameworks for continuous qualification across rapidly evolving AI hardware generations
• Lead root cause analysis (RCA) for systemic issues impacting cluster performance or reliability, driving fixes across stack layers
• Establish qualification gates, acceptance criteria, and release readiness processes aligned with OCI production standards
• Use production telemetry and workload insights to inform validation strategy and proactively identify risk areas
• Influence NIC and system architecture to meet the demands of next-generation AI workloads