QA Engineer | L2/L3 Protocols, Networking , Python , Testing, Manual Testing

• Work with cutting-edge 400G/800G Ethernet fabrics used in AI and GPU data centers.
• Gain exposure to IP routed and VxLAN fabric  and the full NVMeoF stack, from Linux kernel tuning to RoCEv2 fabric automation.
• Hands-on lab environment involving both software-based and hardware-based testing.
• Collaborate with senior architects to compose next-generation lossless Ethernet builds for AI workloads
• Configure and validate Linux/Ubuntu servers for RoCEv2 and NVMeoF traffic generation using 400G NICs.
• Perform system-level tuning including kernel parameters, sysctl, IRQ affinity, NUMA, hugepages, and NIC offloads to optimize RDMA performance.
• Execute NVMeoF traffic tests using software tools (e.g.,IB Perf) and hardware traffic generators (Spirent).
• Capture and analyze fabric metrics such as latency, throughput, packet loss, and congestion response.
• Collaborate with infrastructure teams to ensure lossless Ethernet behavior using PFC and ECN. 
• Configure and validate BGP EVPN/VxLAN-based fabrics for IP-routed data center connectivity.
• Troubleshoot Layer 2/3 networking issues, RoCEv2 packet drops, and flow control problems.
• Collect and analyze NIC counters, RDMA statistics, and telemetry data for performance troubleshooting
• Create and maintain detailed test plans, automation documentation, and configuration templates.
• Generate comprehensive test reports and performance summaries for engineering and management reviews.
• Collaborate cross-functionally with hardware, software, and AI platform teams to enhance fabric performance and reliability.

Minimum Qualifications

• Solid understanding of RDMA, RoCEv2, and NVMe over IP Fabrics (RDMA) architectures.
• Strong hands-on experience with Linux/Ubuntu systems administration and tuning.
• Proficiency in Python scripting and automation frameworks (pytest, paramiko, REST APIs, YAML, Ansible).
• Experience configuring PFC and ECN parameters on Linux servers and network switches for RoCEv2.
• Familiarity with traffic generation tools, and commercial hardware tools (Ixia, Spirent).
• Deep knowledge of network protocols including IP, BGP, VxLAN, TCP congestion control, and QoS.
• Strong debugging skills using Linux utilities (ethtool, tc, sysctl, ibv_devinfo, rdma link, nvme list).
• Experience tuning 400G NICs (MTU, SR-IOV, offloads, NUMA pinning, IRQ balancing).