Senior RDMA Performance Engineer

This is a fully remote position, open to applicants in Poland.

📋 Description

• Set up, troubleshoot, and manage NVIDIA/Mellanox Spectrum switches and ConnectX Network Interface Cards (NICs).

• Deploy and fine-tune congestion control mechanisms tailored for AI workloads, such as RoCEv2, PFC, and ECN.

• Enhance network environments to achieve optimal throughput, reduced latency, and minimal Job Completion Time (JCT).

• Perform performance benchmarking and validation testing on AI-optimized data center fabrics utilizing RDMA and NCCL.

• Utilize AI agents and specialized tools for the automation of network tasks, testing, and network telemetry analysis.

• Create and sustain technical documentation detailing network designs, configurations, and benchmarking outcomes.

• Collaborate within an Agile project team in a startup setting focused on SONiC validation for NVIDIA Spectrum-based AI networking platforms.

⛳️ Requirements

• Over 7 years of professional experience in network engineering.

• Ideally, 3+ years concentrating specifically on data center network design and architecture.

• Practical experience in configuring and troubleshooting BGP, EVPN/VXLAN, LACP, and ECMP.

• Extensive knowledge and hands-on experience with AI-focused data center designs, including Rail-Optimized Design (ROD) and Rail-Unified Design (RUD) topologies.

• Established experience in configuring, managing, and troubleshooting NVIDIA/Mellanox Spectrum switches and ConnectX Network Interface Cards (NICs).

• Profound understanding of RoCEv2, Priority-based Flow Control (PFC), and Explicit Congestion Notification (ECN).

• Experience in tuning switch buffers and implementing Quality of Service (QoS) policies.

• Background in conducting performance benchmarking and validation for AI-ready data center fabrics using RDMA and NVIDIA Collective Communication Library (NCCL).

• Capacity to leverage AI agents/tools for task automation, testing, and analysis.

• Strong skills in technical documentation.

• Proficient in English communication.

• Professional certifications such as CCNP/CCIE Enterprise/DC or JNCIP are valued.

• Familiarity with pyATS, NAPALM, Batfish, Nornir, or NetBox is a plus.

• Exposure to GitLab CI, GitHub Actions, Jenkins, and Infrastructure as Code (IaC) concepts is appreciated.

🏝️ Benefits

• Flexible working hours and options: fully remote, in-office, or hybrid.

• Opportunities for professional growth through internal training sessions and a dedicated training budget.

• Comprehensive onboarding process with a hands-on approach.

• A welcoming environment among professionals passionate about their work.

• The possibility to shift the project you are involved in.