Security Architect, AI Infrastructure

Fluidstack • Austin, TX

2d • $275,000 - $344,000

About The Position

Fluidstack is hiring a Security Architect to design and guide security for frontier AI compute infrastructure. The role owns the attack surface specific to frontier compute, including model-weights custody, training-environment isolation, accelerator and interconnect security, and multi-tenant boundaries under adversarial pressure.

The Security Architect will establish reference architectures and standards for security engineering, review designs for systems connected to training and inference environments—from bare metal and network fabric through schedulers and storage—and ensure workloads are included in the threat model. The role also anticipates emerging developments in frontier compute, such as new accelerators, interconnect generations, confidential computing, and agentic workloads, and prepares security architecture accordingly.

Requirements

• Experience securing AI, ML, or HPC infrastructure where assets were high-value targets, such as lab infrastructure, supercomputing centers, or research environments that attracted nation-state attention.
• Ability to architect isolation across hardware, firmware, hypervisor, kernel, container, network fabric, and physical layers.
• Understanding of AI compute infrastructure as a system, including accelerators, high-speed interconnects, parallel storage, and schedulers, and awareness of how these components challenge classical security assumptions.
• Original work in systems or ML security, published or shipped, and the ability to translate research into architecture engineers can implement.
• Ability to defend design decisions on their merits and revise them when evidence from closer to the hardware warrants it.

Nice To Haves

• Depth in confidential computing and attestation.
• Experience with weights custody or cryptographic key custody at scale.
• A publication record in systems or ML security.

Responsibilities

• Own the security attack surface for frontier compute, including model-weights custody, training-environment isolation, accelerator and interconnect security, and multi-tenant boundaries.
• Design reference architectures and standards for security engineering.
• Architect defensible environments that account for workloads as potential adversaries.
• Conduct security design reviews for systems that touch training and inference environments, from bare metal and network fabric to schedulers and storage.
• Track developments in frontier compute, including new accelerators, interconnect generations, confidential computing, and agentic workloads, and prepare architecture for them.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed