Senior Platform Architect (AI/HPC)

TL;DR

Senior Platform Architect (AI/HPC): Own platform-level architecture for scale-up and scale-out systems as accelerator products grow from single-accelerator to server- and rack-level deployments with an accent on system topology, host integration, memory organization, and performance/scalability/operability. Focus on defining interconnect and fabric requirements, translating workload needs into concrete platform specs, and leading architecture decisions across internal teams and strategic customers.

Location: Hybrid/Remote in Europe (incl. UK); remote from Albania; hybrid in Amsterdam and Eindhoven; on-site in Athens, Bristol, Florence, Leuven, Milan, Paris, and Zurich. Relocation available to Italy (Florence/Milan) or the Netherlands (Amsterdam/Eindhoven). Priority for candidates based in Belgium or Italy.

Company

hirify.global builds a next-generation AI platform and Metis™ AI Platform for high-efficiency AI workloads.

What you will do

• Define platform-level architecture for scale-up and scale-out fabrics across the accelerator product line, from single-node systems to rack-level deployments.
• Specify physical interconnect infrastructure and the architectural trade-offs that impact system-level workload performance.
• Translate workload and performance requirements into system-level architecture and interconnect specifications.
• Define host integration (PCIe, CXL, memory attachment) and platform memory organization.
• Own platform performance, power, cost, scalability, and ensure target performance as systems scale from nodes to clusters.
• Define operability and RAS requirements and lead system architecture decisions with internal teams, partners, and customers.

Requirements

• 5+ years of experience in system, platform, or hardware architecture with a strong track record at server and/or rack scale.
• Strong knowledge of scale-up/scale-out system design, distributed workload mapping, functional partitioning, and fabric/interconnect architecture.
• Hands-on understanding of Ethernet, UALink, optical, and switched-fabric technologies, including system-level performance trade-offs.
• Ability to connect workload characteristics to hardware architecture and quantify end-to-end performance impact.
• Demonstrated ability to lead and collaborate across multidisciplinary teams and interface effectively with senior stakeholders and partners.
• Strong problem-solving skills and a collaborative mindset for building systems at scale.

Nice to have

• Experience architecting AI/HPC accelerator systems and familiarity with distributed training/inference workloads.
• Exposure to data-center-scale deployment.

Culture & Benefits

• Flexible working arrangement with options for office-based work (Belgium, Netherlands, Switzerland, Italy, UK), fully remote work from Europe, or relocation to Italy/Netherlands.
• Attractive compensation package including pension plan, extensive employee insurances, and option to get company shares.
• Open culture supporting creativity and continual innovation, with collaborative ownership and freedom with responsibility.
• Equal opportunity and a diverse, inclusive environment.

Hiring process

• Interviews to assess architecture experience and system/platform design depth.
• Technical discussions focused on platform architecture, interconnect/fabric trade-offs, and RAS/performance ownership.
• Final alignment on architecture collaboration with internal divisions and customer/partner engagement.