Job description

ABOUT FLUIDSTACK We exist to make humanity more free. For most of human history, you farmed or you starved. Technology gave people more time for the things they wanted to do, instead of things they had to do. Powerful AI will be the biggest lever for human choice we've ever built - but only if models are aligned with what humanity actually wants. There are groups building AI who don't share these goals. Whoever deploys frontier compute infrastructure fastest will decide whether AI expands human freedom or shrinks it. We're singularly focused on delivering 10 to 100s of GWs of compute faster than anyone else, rethinking every layer of the stack. We acquire power, design and build data centers, and operate them - with teams spanning hardware and software. Speed and scale are our key differentiators. Come be a part of building civilization-scale infrastructure for AI. We hire people who care deeply about this problem space. If that is you, please apply! HOW WE OPERATE - Be a barrel. Full autonomy. Own things end to end, take on scope without being asked, no permission required to operate outside your core role. - Insane urgency. We drive everything forward as fast as possible. - Reason from first principles. Challenge every assumption. Zero analogy thinking, no egos, the best idea wins. - Love of the game. The frontier of AI is the most interesting problem of our time. We put in long hours at high intensity to push the frontier forward. - Build something that actually matters. If you're going to spend your time, spend it on something that matters to the world. THE INFRASTRUCTURE TEAM Examples of key problems the team is working on - Design the fabrics the frontier trains on. Lossless, non-blocking backend networks for clusters of 100k+ accelerators, re-derived for every new generation of silicon, often before the chip is public. - Multiple fabrics, one system. Frontend, backend, backbone, management, enterprise, and BMS, designed as a single coherent architecture. - Generate the design, don't draw it. Topologies, addressing, BGP/ASN schemas, and golden configs produced from a source-of-truth model, a full site network design in days instead of quarters, correct by construction. ROLE SCOPE - Own the network design lifecycle from customer requirements (GPU shape, workload, scale, tenancy) through deployable, validated architectures for AI training and inference. - Produce topology designs, IP/addressing schemes, routing policy, and fabric configuration specs across front-end, back-end (GPU-to-GPU training fabric), and storage networks. - Adapt architectures to different GPU platforms (NVIDIA, AMD, custom accelerators), form factors, and workload profiles, each with its own rack layout, power envelope, and cabling approach. - Translate logical designs into physical reality: rack elevations, power constraints, structured cabling and fiber budgets, pathway routing, and airflow that affect equipment placement. - Design lossless Ethernet fabrics for RDMA (RoCEv2): PFC, ECN tuning, traffic classes, and congestion management, reasoning about ECMP and collective-communication patterns in distributed training. - Produce HLDs, LLDs, cutsheets, BOMs, cabling matrices, and design decision records, and lead design reviews and reusable reference architectures. WHAT WE'RE LOOKING FOR The below is a starting point. We always make space for exceptional people, so if you don't fit this role exactly, tell us where you would jobs.ashbyhq.com. - You've designed data center network fabrics from requirements through deployment, not just configured them, and can explain the tradeoffs behind every decision. - You have deep L1 to L3 expertise: CLOS/fat-tree topologies, BGP, EVPN/VXLAN, and the fundamentals underneath them. - You design lossless RDMA (RoCEv2) fabrics and understand congestion management at training scale. - You reason from first principles through novel design challenges rather than pattern-matching to one reference architecture. - You partner across Hardware, DC Operations, ICT, Software, and Validation so designs are buildable and operationally sound. - Bonus: Source-of-truth-driven design generation. Multi-vendor GPU platform integration. Large-cluster (100k+ accelerator) experience. SALARY & BENEFITS - Competitive total compensation package (salary + equity). - Retirement or pension plan, in line with local norms. - Health, dental, and vision insurance. - Generous PTO policy, in line with local norms. We are committed to pay equity and transparency. Fluidstack is an Equal Employment Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, sexual orientation, gender identity, disability and protected veterans’ status, or any other characteristic protected by law. Fluidstack will consider for employment qualified applicants with arrest and conviction records pursuant to applicable law. You will receive a confirmation email once your application has successfully been accepted. If there is an error with your submission and you did not receive a confirmation email, please email careers@fluidstack.io with your resume/CV, the role you've applied for, and the date you submitted your application-- someone from our recruiting team will be in touch.

Requirements

Must-haves

Be a barrel. Full autonomy. Own things end to end, take on scope without being asked, no permission required to operate outside your core role.
Insane urgency. We drive everything forward as fast as possible.
Reason from first principles. Challenge every assumption. Zero analogy thinking, no egos, the best idea wins.
Love of the game. The frontier of AI is the most interesting problem of our time. We put in long hours at high intensity to push the frontier forward.
Build something that actually matters. If you're going to spend your time, spend it on something that matters to the world.
Examples of key problems the team is working on
Design the fabrics the frontier trains on. Lossless, non-blocking backend networks for clusters of 100k+ accelerators, re-derived for every new generation of silicon, often before the chip is public.
Multiple fabrics, one system. Frontend, backend, backbone, management, enterprise, and BMS, designed as a single coherent architecture.
Generate the design, don't draw it. Topologies, addressing, BGP/ASN schemas, and golden configs produced from a source-of-truth model, a full site network design in days instead of quarters, correct by construction.
Own the network design lifecycle from customer requirements (GPU shape, workload, scale, tenancy) through deployable, validated architectures for AI training and inference.
Produce topology designs, IP/addressing schemes, routing policy, and fabric configuration specs across front-end, back-end (GPU-to-GPU training fabric), and storage networks.
Adapt architectures to different GPU platforms (NVIDIA, AMD, custom accelerators), form factors, and workload profiles, each with its own rack layout, power envelope, and cabling approach.
Translate logical designs into physical reality: rack elevations, power constraints, structured cabling and fiber budgets, pathway routing, and airflow that affect equipment placement.
Design lossless Ethernet fabrics for RDMA (RoCEv2): PFC, ECN tuning, traffic classes, and congestion management, reasoning about ECMP and collective-communication patterns in distributed training.
Produce HLDs, LLDs, cutsheets, BOMs, cabling matrices, and design decision records, and lead design reviews and reusable reference architectures.
The below is a starting point. We always make space for exceptional people, so if you don't fit this role exactly, tell us where you would https://jobs.ashbyhq.com/fluidstack/05c2e69c-42f9-4fcb-9cf0-a467aaf98f1c.
You've designed data center network fabrics from requirements through deployment, not just configured them, and can explain the tradeoffs behind every decision.
You have deep L1 to L3 expertise: CLOS/fat-tree topologies, BGP, EVPN/VXLAN, and the fundamentals underneath them.
You design lossless RDMA (RoCEv2) fabrics and understand congestion management at training scale.
You reason from first principles through novel design challenges rather than pattern-matching to one reference architecture.
You partner across Hardware, DC Operations, ICT, Software, and Validation so designs are buildable and operationally sound.
Competitive total compensation package (salary + equity).
Retirement or pension plan, in line with local norms.

Nice-to-haves

Bonus: Source-of-truth-driven design generation. Multi-vendor GPU platform integration. Large-cluster (100k+ accelerator) experience.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

Dental

Vision

Pto