Staff Security Software Engineer, AI Security

RDQ426R108

This role is open to candidates in the US (any location)

About the Team

The AI Security team at Databricks sits at the frontier of securing the AI/ML services in the Databricks platform. As we ship AI capabilities at the leading edge of the industry, including Agent Bricks, the Genie suite, AI Model Serving, MLflow, and Unity AI Gateway, the AI Security team ensures these systems are designed, built, and operated securely. Our work also extends to securing our own usage of AI: building the right guardrails that enable Databricks employees to innovate and deliver securely.

The team combines offensive security depth with AI/ML engineering knowledge to identify novel threats, build scalable defenses, and influence how AI products are architected from the ground up. We lead AI Red Team exercises, build security tooling for AI workloads, and partner directly with AI Product teams to embed security into the development lifecycle.

---

The Role

As a Staff Security Software Engineer on the AI Security team, you are a senior technical leader who sets the standards for how Databricks secures its AI and ML capabilities. You combine deep offensive security expertise with practical knowledge of AI/ML systems to identify and drive resolution of the most significant security risks in Databricks' AI platform.

You lead AI red team engagements against production AI systems, conduct security architecture reviews for complex, multi-system AI features, and build the tooling and frameworks that scale the team's impact. You are a subject matter expert in at least two AI security domains and you operate with significant autonomy- driving cross-team remediation, setting technical standards, and mentoring teammates in both offensive techniques and secure AI design.

---

The Impact You Will Have

AI Red Team & Adversarial Testing

• Lead AI red team engagements against Databricks' production AI systems, including Foundation Model APIs, Genie and natural language query systems, Model Serving infrastructure, MCP-connected agents, and RAG pipelines
• Design and execute adversarial attack scenarios: prompt injection, jailbreaking, memory poisoning, cross-tenant data leakage in multi-tenant serving, and sandbox bypasses
• Develop proof-of-concept exploits for AI-specific vulnerability classes and perform variant analysis to identify the full scope of exposure across the AI platform
• Contribute to the evolution of the Databricks AI Security Framework (DASF), maintaining and extending the risk taxonomy, control library, and testing methodology as AI capabilities evolve

AI Product Security & Architecture Reviews

• Lead comprehensive security architecture reviews for complex AI features: threat modeling agentic workflows, RAG