Senior Offensive AI Security Researcher

About the role

Salesforce is seeking a senior researcher to lead novel offensive AI security research across its AI ecosystem. The role focuses on uncovering systemic vulnerabilities in generative AI, retrieval‑augmented generation, and autonomous agent workflows before they can be exploited.

Key responsibilities

• Scope, design, and execute advanced adversarial research targeting AI threat vectors such as indirect prompt injection, tool abuse, RAG system poisoning, data extraction, and supply‑chain compromises.
• Develop and operationalize new attack techniques, publishing internal and external threat intelligence and vulnerability analyses.
• Provide an offensive security perspective to guide AI architecture, shape internal threat models, and contribute to Salesforce’s Generative AI Security Standard.

Required profile

• 5+ years in offensive security roles (vulnerability research, penetration testing, application security, or security engineering).
• 2+ years hands‑on experience auditing or breaking ML/AI systems with deep expertise in LLM attack surfaces.
• Proven ability to lead complex security initiatives and communicate high‑stakes technical risks to engineers and executives.

Required skills

• Python for exploit proof‑of‑concept development and data analysis.
• Deep technical understanding of modern AI design patterns, including Retrieval‑Augmented Generation, autonomous agentic workflows, and AI pipelines.
• Experience with ML frameworks and tools such as PyTorch, Hugging Face, LangChain, and vector databases.