Make RDMA First-Class: Integrate RDMA/RoCE/InfiniBand capabilities into our inference stack.
Optimize Distributed Inference: Implement and tune networking layers for Disaggregated KV Cache Offload and WideEP.
Enable Serverless-Grade Startup Speeds for LLMs: Work with checkpointing and storage for sub-10-second startup for models.
Deep-Dive into Hardware: Validate networking performance on bleeding-edge clusters and write acceptance tests.
Build Observability: Design tools to visualize packet flow and diagnose distributed system behaviors.
Optimize Kernels: Work with communication libraries (NCCL, NVSHMEM) and write custom kernels to overlap compute and data transfer.

Requirements

Deep experience with high-performance networking protocols (InfiniBand, RoCE v2) and understand the physics of data movement.
Fluent in C++ or Python, with the ability to bridge the gap between high-level logic and hardware.
Deep understanding of the memory hierarchy in modern NVIDIA architectures (H100/Blackwell) and know how to optimize for it.
Experience with NCCL, NVSHMEM, and UCX is highly preferred.
Experience with GPUDirect Storage (GDS) or high-performance filesystems like Weka or 3FS.
Familiarity with TensorRT-LLM, vLLM, or Sglang is a plus.
Experience running low-level benchmarks to "qualify" new hardware clusters.

Benefits

Competitive compensation, including meaningful equity.
100% coverage of medical, dental, and vision insurance for employee and dependents
Generous PTO policy including company wide Winter Break (our offices are closed from Christmas Eve to New Year's Day!)
Paid parental leave
Company-facilitated 401(k)
Exposure to a variety of ML startups, offering unparalleled learning and networking opportunities.

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills & Tools

C++PythonRDMARoCEInfiniBandNCCLNVSHMEMUCXGPUDirect StorageTensorRT-LLM