SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salesforce is seeking a Senior Offensive AI Security Researcher to lead novel adversarial research across its AI ecosystem, uncovering critical vulnerabilities in emerging AI threat vectors before exploitation. This role combines academic research with practical offensive methodologies to establish new tactics and techniques targeting complex AI paradigms, including indirect prompt injection, tool abuse and excessive agency, RAG system poisoning, data extraction, and model/data supply chain compromises.
You will scope, design, and execute advanced adversarial research applying risk-based approaches to uncover novel threats. Working as a trusted partner to engineering and architecture teams, you'll provide offensive security perspectives that shape internal threat models, guide AI architecture decisions, support corporate governance, and directly contribute to Salesforce's Generative AI Security Standard.
Key responsibilities include leading novel AI security research initiatives, innovating and operationalizing AI attack techniques, guiding security strategy and standards, and disseminating research through internal and external threat intelligence, vulnerability analyses, and security publications.
Required qualifications: 5+ years in offensive security roles (vulnerability research, penetration testing, application security, or security engineering); 2+ years hands-on experience auditing, breaking, or researching ML/AI systems with deep expertise in LLM attack surfaces and frameworks (OWASP Top 10 for LLMs, MITRE ATLAS); deep technical understanding of modern AI design patterns including RAG, autonomous agentic workflows, model context protocols, and AI pipelines; high proficiency in Python for exploit development and technical evaluation; proven track record leading complex security initiatives and communicating critical risks to research engineers and executives.
Preferred: Advanced degree (MS/PhD) in Computer Science, Machine Learning, Cybersecurity, or related field; public portfolio of offensive security research including conference presentations (Black Hat, DEF CON, NeurIPS), published papers, CVEs, advisories, or open-source contributions; experience auditing modern AI development ecosystems and tools (PyTorch, Hugging Face, LangChain, vector databases, open-weight models).
