LinkedIn, Netflix, DoorDash reveal how they actually run AI in production

QCon AI New York 2026 has released its schedule, featuring engineers from major tech companies like LinkedIn, Netflix, DoorDash, and Figma. They will share detailed case studies on the real-world architecture and operational challenges of running AI systems at scale, covering topics like agent authorization, evaluation pipelines, model serving, and AI-assisted software delivery.

For developers and engineers moving beyond initial AI experiments, these talks offer a rare look into the trenches of production AI. Speakers will address critical questions: where to enforce agent permissions, how to connect offline evaluations to live system behavior, and how to manage growing inference workloads. The key takeaway is that scaling AI requires solving hard engineering problems around security, reliability, and evaluation—less about the models themselves and more about the robust systems that host them.