Biography
Antony Alangaram is a Principal Engineer and Enterprise Architect specializing in modern software architecture, production engineering, cybersecurity, and cloud resilience. With deep expertise in scaling enterprise systems, technical architecture design, and modernizing complex cloud ecosystems, Antony focuses on turning emerging technologies into secure, production-grade solutions. An author and speaker on secure AI architectures and scalable infrastructure, Antony advocates for zero-trust design principles across non-deterministic systems.
Abstract
Title: Secure AI Systems and Applications: Building – Resilient, Zero-Trust Architectures for Production Environments
Abstract: As software engineering shifts from fixed, deterministic logic toward probabilistic, AI-driven architectures, securing production applications presents unprecedented challenges. Traditional cybersecurity models—such as perimeter firewalls, static role-based access control (RBAC), and web application firewalls (WAFs)—assume predictable code execution paths. They fall short when defending applications powered by Large Language Models (LLMs) and probabilistic software engines, where data and instructions inhabit the same unsegregated execution channel. This exposes software systems to novel threat vectors, including direct/indirect prompt injection, context data exfiltration, model supply chain poisoning, and unauthorized API execution.
This presentation outlines an engineering framework for designing and deploying Secure AI Systems and Software Applications. Built upon Zero Trust principles and secure-by-default architecture, this session demonstrates how to construct multi-layered technical controls across the entire application lifecycle—from data ingestion and context isolation to runtime guardrails, dynamic authorization, and continuous observability.
