---
title: "Principal Sovereign GPU Cloud Infrastructure Architect"
slug: "principal-sovereign-gpu-cloud-infrastructure"
id: "60e4d961-c762-4d20-a1ac-29446a0d12b7"
department: "AI Infrastructure"
department_code: "INF"
location: "London, UK"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "£130k – £175k (annual package)"
is_remote: false
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure"
markdown_url: "https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:27:32.396Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Principal Sovereign GPU Cloud Infrastructure Architect

## Position Telemetry

- **Reference:** `PRINCIPAL-SOVEREIG`
- **Practice / Sector:** AI Infrastructure (`INF`)
- **Location:** London, UK
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** £130k – £175k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is one of Europe's largest and fastest-growing AI infrastructure unicorns ($1B+ capital raised), headquartered in London and building vertically integrated, liquid-cooled sovereign GPU data centers from the ground up. They provide frontier model laboratories, national sovereign AI programmes, and global enterprises with dedicated multi-thousand-GPU training clusters and managed Kubernetes AI clouds free from hyperscaler virtualization tax.

They have engaged Talent247 to appoint a Principal Sovereign GPU Cloud Infrastructure Architect in London. Reporting into senior infrastructure leadership as a high-leverage technical authority, you will define the end-to-end bare-metal-to-Kubernetes architecture governing how tens of thousands of next-generation accelerators (HGX H200, GB200 NVL72) are provisioned, networked, isolated for strict multi-tenancy, and kept at >98% goodput.

KEY RESPONSIBILITIES
• Define and evolve the reference compute, storage, and fabric architecture for multi-site bare-metal Kubernetes and Slurm GPU superclusters scaling from thousands to tens of thousands of accelerators.
• Architect zero-touch bare-metal lifecycle automation—spanning PXE/network boot, BMC/Redfish firmware orchestration, immutable Linux/driver imaging, automated burn-in validation, and silent-data-corruption quarantine.
• Design rail-optimized 400G/800G InfiniBand and RoCEv2 lossless fabrics alongside L2/L3 cryptographic tenant isolation, SR-IOV/GPUDirect RDMA passthrough, and sovereign enclave compliance boundaries.
• Partner with distributed storage leads to engineer multi-petabyte parallel filesystem and caching tiers (Lustre, WEKA, VAST, Ceph) delivering 20–50 GB/s per node for checkpointing and weight loading.
• Lead cross-team architectural reviews, hardware vendor co-design sessions, and root-cause analyses on complex hardware-kernel-network interactions.
• Mentor staff and senior engineers, establishing reusable operators and operational standards across global data center expansions.

TECHNICAL ENVIRONMENT & OFFERING
Their stack spans bare-metal Linux, Kubernetes custom controllers (Go/Rust), Slurm, NVIDIA GB200 NVL72 / HGX H200, InfiniBand/SONiC/EVPN, Redfish/BMC automation, WEKA/VAST parallel storage, and Terraform. Based hybrid in central London, our client offers £130,000–£175,000 GBP base salary plus substantial unicorn equity, executive benefits, and UK visa/relocation support.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 8+ years of infrastructure, HPC, neocloud, or hyperscaler systems architecture experience, including proven technical leadership over large-scale bare-metal GPU or supercomputing estates.
• Deep hands-on mastery of NVIDIA multi-node GPU architectures (HGX H100/H200, NVLink/NVSwitch, GB200 NVL72, MIG, DCGM) and GPUDirect RDMA networking (InfiniBand or RoCEv2).
• Strong systems engineering track record building Kubernetes bare-metal operators/controllers (Cluster API, Metal3, Kueue, Volcano) and/or Slurm orchestration in Go, Rust, or Python.
• Thorough command of Linux kernel/driver internals, BMC/Redfish hardware lifecycle tooling, and multi-tenant cloud security isolation.
• Exceptional architectural communication skills across data center hardware, networking, security, and ML research stakeholders.

PREFERRED / BONUS EXPERIENCE
• Experience commissioning liquid-cooled GB200 rack-scale clusters or designing sovereign/government-accredited cloud enclaves.
• Hands-on familiarity with NCCL collective profiling and parallel AI storage fabrics (WEKA, VAST, Lustre).

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/principal-sovereign-gpu-cloud-infrastructure or via HTTP `POST https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/principal-sovereign-gpu-cloud-infrastructure/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
