Role OverviewDevelop and execute tests for collective communication operations on custom networking silicon, validating the ASIC's ability to enable large-scale distributed AI training workloads.
What You Will Do
Develop test suites, write RDMA verbs-level tests, validate multi-node collective communication patterns, and integrate tests into CI/CD pipelines.
Why It Might Be a Fit
Must have deep understanding of RDMA operations, RoCE v2, collective communication operations, and Ethernet/L2 networking. Experience with pre-silicon validation environments, open-source RDMA/networking projects, and build systems at scale is a plus.
Requirements
- RDMA (Remote Direct Memory Access) — Deep understanding of RDMA operations: READ, WRITE, SEND, RECEIVE; Queue Pairs (QPs), Completion Queues (CQs), Memory Regions (MRs), Protection Domains (PDs)
- RoCE v2 — Understanding of RDMA over Converged Ethernet protocols, transport-level behavior, and conformance requirements
- Collective Communication Operations — AllReduce, AllGather, ReduceScatter, AllToAll; ring/tree algorithms; understanding of how collectives map to network traffic patterns
- Ethernet / L2 Networking — Layer 2 fundamentals, MTU, multiport networking, VLANs
- PCIe Architecture — PCIe endpoint/switch topology, Gen5/Gen6, BAR regions, MSI-X interrupts, SR-IOV, multi-function devices
- NCCL / Communication Libraries — Familiarity with NVIDIA Collective Communications Library or equivalent; understanding of how training jobs use collectives over RDMA NICs
Benefits
- medical, dental, vision, pharmacy, life, accidental death & dismemberment, and disability insurance
- employee assistance program
- 401(k) retirement plan
- 10 days of paid time off per year
- 10 paid holidays per year
]]>