Modern AI isn't just a model, but rather a mostly a messy, multi-step pipeline. When you chain together ingestion, tool calls, and long-term memory, you aren't just building an agent—you’re building a new class of vulnerabilities. Traditional security looks at the model; but you need to also be looking at the spaces between the steps. In this session, Vinay will move beyond basic prompt injection to explore how risks cascade through an entire workflow. He’ll break down: cascading injections, privilege escalation via tooling (when your AI’s "helpful" integration becomes a back door), and...
Vinay Mishra

Vinay Mishra is a global engineering leader and Vice President at Flagstar, where he bridges the gap between complex technical architecture and bottom-line business outcomes. With over 21 years of experience across the US, EMEA, and APAC, Vinay has spearheaded large-scale digital transformations in highly regulated sectors, including finance, healthcare, and manufacturing. A Doctoral (D.Eng.) researcher/student at The George Washington University, he blends academic rigor with hands-on expertise in applications, infrastructure and techno-management fields. Vinay is a PMP® and CSP-PO® certified professional, and a strategic advisor dedicated to building resilient, innovation-ready engineering organizations.