Senior Network Engineer

Responsibilities

• Design and implement simulation models for networking hardware and infrastructure, leveraging Discrete Event Simulation tools (e.g., Omnet++, NS3, QualNet, DONS).
• Develop and improve the Discrete Event Simulation (DES) software solution for data center networks, focusing on high-speed networks that feature very high throughput and very low latency
• Identify congestion bottlenecks and optimal link placements for multi-GPU clusters
• Integrate hardware and software components to enable seamless operation of distributed networking systems and AI-based applications
• Stay informed about cutting-edge advancements in networking (e.g., RDMA, Ultra Ethernet, Software-Defined Networking, congestion control), AI technologies, distributed systems, and datacenter hardware. Apply these advancements to drive innovation in simulations
• Implement datacenter switch behaviors in NS3 (buffer management, packet scheduling, priority queues) to accurately model vendor-specific hardware
• Model LLM training (e.g., GPT-scale models) in Astra-sim across 100s-1000s of GPUs, predicting training iteration time based on network topology and collective communication patterns
• Develop and evaluate network topologies, collective operations (MPI/NCCL/xCCL libraries), and distributed networking approaches to optimize performance and scalability
• Research and implement emerging technologies like Ultra Ethernet, in-network computing, or congestion control innovations in the simulator
• Create and maintain comprehensive technical documentation, including design specifications, test plans, user manuals, and process workflows.

Languages