Network Engineer Design and Engineering
FluidstackVisit Fluidstack website
AI infrastructure company that builds and operates large-scale compute and data-center infrastructure for frontier AI workloads.
New York City, United States
Funding history
About Fluidstack
Fluidstack deploys AI compute infrastructure, including custom data centers and large-scale compute capacity, for AI labs, governments, and enterprises.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will take AI training and inference network designs from customer requirements to validated architectures. You will produce topology, addressing, routing, fabric, and physical infrastructure designs, design lossless RDMA Ethernet fabrics, create technical design documents, and lead design reviews.
Requirements
- Design data-center network fabrics from requirements through deployment
- Apply L1 to L3 networking knowledge, including CLOS or fat-tree topologies, BGP, and EVPN/VXLAN
- Design lossless RDMA and RoCEv2 fabrics with congestion management
- Collaborate with hardware, data-center operations, ICT, software, and validation functions
- Apply source-of-truth design generation, multi-vendor GPU-platform integration, or large-cluster networking experience
Responsibilities
- Own the network-design lifecycle from requirements through validated architecture
- Produce topology, addressing, routing-policy, and fabric-configuration designs
- Adapt designs for GPU platforms, form factors, and workloads
- Translate logical designs into rack, power, cabling, pathway, and airflow requirements
- Design lossless Ethernet fabrics for RDMA and RoCEv2
- Produce HLDs, LLDs, cutsheets, BOMs, cabling matrices, and design decision records
- Lead design reviews and develop reusable reference architectures
Benefits
- Equity
- Retirement or pension plan
- Health insurance
- Dental insurance
- Vision insurance
- Generous PTO policy