Aviral Srivastava

Aviral Srivastava is a Security Engineer at Amazon focused on AI application security and building enterprise-secure AI systems. His primary research area is offensive security for AI targeting agentic systems and reinforcement learning with red-team methods that surface tool-use abuse and second-prompt risks. His work has appeared in venues such as NeurIPS, ICLR, and ACM. An RSAC Security Scholar, Aviral has spoken at CypherCon, CactusCon, and multiple BSides events, and received “Cybersecurity Innovator of the Year” at BSides Bangalore. He is the Author of "ML for Hackers." Aviral earned an MS in Cybersecurity Analytics and Operations from Penn State University, where he worked on adversarial ML, symbolic execution, and automated exploit generation.