Why it matters
Production-readiness for AI agents is where most teams hit their first real cost wall — scaling, observability, and fallback routing in production environments are qualitatively different from prototype work, and this event targets exactly that gap.
The tokenmaxxing angle
Five speakers from different companies showcasing production agent deployments on AWS gives attendees a cross-section of how organizations are instrumenting and controlling inference spend at scale — including AWS Bedrock routing and cost observability patterns.
From the organizers
David Potes (Sr. Manager, Data & AI at AWS), Victor Wang (Deepgram), John Wang (Saviynt VP), Kevin Locke (4Minds), and Sachin Mohan (Zendesk) present at Amazon AWS, 525 Market St, San Francisco, CA on June 24, 4:30–7:30 PM PDT.