As AI agents move from prototypes to production, developers face a growing challenge: reliability and trust. It’s no longer enough for an agent to work, it must be transparent, auditable, and resilient.
In this talk, Disleve Kanku (AI Engineer at Dana Farber Cancer Institute & Founder of Blueprint Intel) unpacks the technical and cultural frameworks behind building trustworthy AI systems. He’ll walk through the architecture of agentic pipelines, from real-time feedback loops and event-driven orchestration to automated validation and continuous monitoring and show how developers can embed “trust layers” directly into their codebase.
You’ll learn how to design for explainability, align agents with human intent, and create reproducible deployment practices that scale safely across domains. Whether you’re an MLOps engineer or a startup developer exploring GPT-based automation, you’ll leave with a practical blueprint for building AI that both works and earns trust.