Staff HPC Network Architect
Lambda Labs · San Jose Office (First St) · San Francisco Office (Fremont St)
CHECKED 3H AGO
Infrastructureremotestaff$284k–$420k
At a glance
Highlights
- Architect ultra-low-latency networks powering AI cloud infrastructure
- Design topologies for large-scale GPU clusters and storage backends
- Define scalability roadmaps for multi-site cloud environments
What they offer
- Equity compensation
- Health, dental and vision coverage for dependents
- Wellness and commuter stipends
- 401(k) with 2% company match
- Flexible paid time off
Ideal candidate
Veteran data-center network architect with 7+ years shaping HPC or AI clouds. Self-starter with strong ownership who mentors others and thrives in ambiguity to push ultra-low-latency fabrics.
Requirements
- 7+ years architecting high-performance data center networks for HPC, AI/ML or cloud
- Deep expertise with InfiniBand and advanced Ethernet fabrics including RoCE and RDMA
- Strong understanding of switching architectures, PFC/ECN, QoS, VXLAN and EVPN
- Experience designing low-latency high-throughput GPU and storage data paths
- Expertise with BGP fabric design including eBGP underlays and MP-BGP EVPN
- Experience with SDN, automation APIs and network telemetry
- Strong optical networking competency including transceivers, fiber and WDM
- Track record building resilient fault-tolerant architectures with high availability
- remote policy
- 4 days in office, work from home on Tuesdays
Lab profile
About Lambda Labs
The Superintelligence Cloud delivering GPU compute infrastructure for AI researchers and enterprises.
What they build
- GPU cloud infrastructure
- On-demand GPU instances
- Managed Kubernetes for AI workloads
Known for
- AI cloud infrastructure at scale
- HPC platform and hardware engineering
- Data center operations and networking
Key facts
- Serves tens of thousands of customers
- Offices in San Francisco, San Jose and Bellevue
- Data center in Quebec City, Canada