Senior Security Architect, Cloud Authentication and Authorization

NVIDIA•Us, CA

3d•Onsite

About The Position

NVIDIA is seeking a Senior Security Architect, Cloud Authentication and Authorization to join their groundbreaking team in Santa Clara, CA. This role involves guiding the architectural vision for identity and authorization systems, collaborating with world-class experts to deliver impeccable solutions. The position offers the opportunity to drive meaningful change in the next generation of computing through innovation and cybersecurity expertise.

Requirements

8+ years experience in cybersecurity, security architecture, cloud security, IAM, application security, product security, platform security, infrastructure security, or security engineering for distributed systems.
Extensive knowledge in cloud authentication, authorization, IAM, workload identity, agent identity, non-human identity, or identity architecture, combined with hands-on experience in developing, managing, deploying, or assuming direct responsibility for authentic security controls.
Bachelor’s degree in Engineering, Cybersecurity, Data Engineering, or a related technical field, or equivalent experience.
Proficiency in authentication and authorization protocols and frameworks, such as OIDC, OAuth 2.0, SAML, federation, delegation, token exchange, token scope, issuer and audience boundaries, consent, mTLS, certificate-backed identity, prioritized access, and associated technologies.
Direct involvement in handling workload and agent identities, covering attestation processes, Zero Trust Architecture concepts, short-lived credentials, and temporary identities.
Experience developing authorization boundaries for distributed systems, including fine-grained authorization patterns, control points, prioritized delegation, model/data/tool access controls, sensitive-action approval, and execution boundaries.
Proficiency with identity and certificate lifecycle management, including enrollment, provisioning, scope definition, prioritized issuance, renewal, rotation, revocation, expiration, auditability, deprovisioning, lifecycle automation, and awareness of crypto-agility and post-quantum cryptography implications.
Hands-on understanding of AI security risks combined with adequate proficiency in AI-enabled systems to assess timely injection, data exfiltration, unsafe tool use, overbroad authorization, and loss of human accountability.
Strong foundational cybersecurity judgment, including threat modeling, architecture review, risk analysis, practical mitigation development, clear communication of assumptions, partner-team alignment, and follow-through through implementation, verification, documentation, and closure.

Nice To Haves

Experience crafting or adopting workload identity systems such as SPIFFE/SPIRE, workload identity federation, service mesh identity, policy engines, or attestation-backed identity provisioning.
Extensive knowledge of autonomous agent identity, delegated authority, token exchange, prioritized credentials with limited scope, certificate-backed identities, identity-aware policy controls, or ownership models for human, workload, service, and agent identities.
Experience crafting controls for AI agent tool use, such as per-tool authorization, policy controls points, approval gates, egress restrictions, connector-scoped credentials, or emergency disablement of compromised agents.
Background with crafting security architecture for enterprise connectors, AI assistants, tool integrations, automation systems, sensitive-action approvals, or cross-system authorization boundaries.
Experience reducing or eliminating static credentials through workload identity, short-lived credentials, certificate lifecycle improvements, auditable service identity, or automated revocation and rotation.

Responsibilities

Outline the security architecture strategy for cloud authentication, authorization, workload identity, and agent identity across NVIDIA cloud platforms, AI-enabled systems, enterprise connectors, services, and automation.
Outline processes for establishing, linking, authorizing, delegating, auditing, and retiring human, workload, service, and autonomous agent identities, including attestation-supported identity issuance and certificate-based or temporary credentials.
Develop authorization and delegation frameworks for AI agents and enterprise connectors, encompassing consent, token exchange, prioritized authority, sensitive-action approval, revocation, and protections against confused-deputy behavior.
Lead architecture reviews and threat modeling for high-risk identity and access flows, turning ambiguous scenarios into practical controls that engineering teams can build and verify.
Establish identity lifecycle, telemetry, and emergency-disablement patterns for token issuance, policy decisions, privilege elevation, tool invocation, data access, credential rotation, grant revocation, and compromised or untrusted identities.
Convert emerging AI security risks into authentication, authorization, audit, and execution-boundary requirements.
Partner with identity, cloud, platform, application, AI security, governance, detection, and incident response teams to align architecture decisions with risk strategy and operational reality.
Build reusable architecture patterns, decision records, exception criteria, and implementation mentorship, staying engaged through adoption, validation, and residual-risk closure.