AI/LLM Security Testing & Red Teaming Lead (Agentforce Platform)
TextRed Teaming
Led AI/LLM Security Testing frameworks for Salesforce's Agentforce platform, evaluating resilience against various attacks. Designed workflows to assess and triage AI-specific vulnerabilities in text-based interactions. Drove a faster triage cycle for issues related to prompt injection, jailbreaks, and data leakage.
• Developed tailored evaluation scenarios for LLM/AI prompt resistance analysis.
• Conducted red teaming for AI system vulnerability, focusing on generative text outputs.
• Coordinated vulnerability report triage for AI labeling and system improvement.
• Automated portions of vulnerability workflow leveraging internal/proprietary tools.
Led AI/LLM Security Testing frameworks for Salesforce's Agentforce platform, evaluating resilience against various attacks. Designed workflows to assess and triage AI-specific vulnerabilities in text-based interactions. Drove a faster triage cycle for issues related to prompt injection, jailbreaks, and data leakage. • Developed tailored evaluation scenarios for LLM/AI prompt resistance analysis. • Conducted red teaming for AI system vulnerability, focusing on generative text outputs. • Coordinated vulnerability report triage for AI labeling and system improvement. • Automated portions of vulnerability workflow leveraging internal/proprietary tools.
2025 - Present