GPU Networking Engineer — RDMA/NVLink for Large-Scale AI

Bison; San Francisco (CA); On-site; USD 350,000 - 475,000; Full time

Benefits offered by this job

Health insurance

Dental benefits

Vision benefits

Unlimited PTO

Parental leave

Relocation support

Job summary

Thinking Machines is hiring a network engineer to own the lowest layers of the network stack for large-scale training and inference. You will ensure interconnect reliability across GPU fabrics, including RDMA/RoCE, NVLink, and NVSwitch, while building instrumentation and tooling for faster debugging.

You’ll work cross-stack with researchers and cloud providers, translating complex networking challenges into reliable, scalable systems.

Qualifications

• Bachelor’s degree or equivalent in CS, engineering, or similar.
• Proficiency in at least one backend language (Python or Rust).
• Experience operating large-scale clusters and container orchestration systems (Kubernetes or Slurm).
• Comfort operating across the stack and owning projects end-to-end.
• Thrive in a highly collaborative environment with cross-functional partners.
• Bias for action and initiative to work across stacks and teams to ship.

Responsibilities

• Reason about and validate GPU network fabric design across deployments.
• Debug RDMA / RoCEv2 across NIC vendors; diagnose NCCL/PFC/ECN issues.
• Own NVLink / NVSwitch interconnect and health, errors, and integration with collectives.
• Build host-level network instrumentation and dashboards using Linux tooling.
• Triage cross-cloud fabric quirks across NICs, drivers, kernels, switches, and workloads.
• Drive escalations with cloud-provider networking teams to resolution.

Skills

Backend development

System design

Collaboration

Initiative

Education

Bachelor’s degree or equivalent in CS/Engineering

Tools

Kubernetes

Slurm

Linux

NCCL

NVLink/NVSwitch

Job description

Thinking Machines is hiring a network engineer to own the lowest layers of the network stack for large-scale training and inference. You will ensure interconnect reliability across GPU fabrics, including RDMA/RoCE, NVLink, and NVSwitch, while building instrumentation and tooling for faster debugging.

You’ll work cross-stack with researchers and cloud providers, translating complex networking challenges into reliable, scalable systems.