As a Security Architect, you will shape the foundation of security for advanced AI silicon and computing systems. Your work will center on developing and scaling security architectures that ensure integrity, confidentiality, and resilience across hardware, firmware, and software layers. You'll play a key role in advancing secure AI infrastructure tailored for enterprise and regulated environments.
Key Responsibilities
- Lead research and development of cutting-edge security technologies for AI silicon, including hardware root-of-trust, secure boot, and confidential computing.
- Define and maintain comprehensive security architecture specifications that balance performance, efficiency, and protection.
- Collaborate with silicon, firmware, software, and systems teams to embed security into every phase of the product lifecycle.
- Partner with leadership to align security strategy with market demands and competitive positioning.
- Develop threat models and security protocols for AI accelerators and large-scale cluster deployments.
- Drive improvements in secure manufacturing, provisioning, and supply chain security practices.
- Stay engaged with evolving industry standards such as Caliptra and contribute to external security initiatives.
Qualifications
Applicants should hold a BS, MS, or PhD in Computer Science, Electrical Engineering, or a related field, along with over a decade of experience in hardware security architecture—ideally focused on AI accelerators, GPUs, or high-performance SoCs. You must have a demonstrated history of translating security architecture into real-world product implementations.
Required expertise includes hardware root-of-trust, secure boot, firmware security, and cryptographic applications in silicon. Practical experience with security frameworks like Caliptra is essential. You should be able to define micro-architectural security features such as memory protection, isolation, and key management, and translate high-level requirements into technical specs across domains.
Strong analytical, debugging, and communication skills are critical. You must thrive in cross-functional settings, adapt quickly to new domains, and work independently with minimal oversight.
Preferred Background
- 15+ years in hardware security, particularly in AI, supercomputing, or hyperscale systems.
- Experience with confidential computing standards and system-level security integration involving BMC, Docker, and Kubernetes.
- Familiarity with commercial security IP and optical switch security considerations.
- Track record of evaluating trade-offs between security, performance, and total cost of ownership in AI workloads.
- Leadership in cross-functional initiatives or industry security groups.
