Dive into the future of AI and MLOps with cutting-edge talks from leading industry leaders on the online stage.
🚀 We’re kicking off the 7th Annual MLOps World | GenAI Summit with a free virtual day on November 16, bringing the global AI practitioner community together before we meet in Austin.
Spend the day exploring what’s happening across the AI landscape, from building and evaluating agentic systems to scaling infrastructure, improving reliability and navigating the realities of putting AI into production.
Whether you’re an AI Engineer, Agent Builder, Solution Architect, infrastructure team member, Full-Stack Developer or any type of technologist looking to raise your AI game, expect practical insights, fresh perspectives and real-world lessons from the people building and deploying AI every day.
🤔 DYK? MLOps World is built by AI practitioners, for AI practitioners, with a program shaped around the topics and challenges that matter most to the community.
KEEP THE MOMENTUM GOING IN AUSTIN | NOVEMBER 17–18
Then, we’re taking things offline and heading to Austin.
Join AI practitioners, technologists, enterprise teams, startups and industry leaders for two days of expert-led sessions, big ideas, technical conversations and plenty of opportunities to connect.
With 16 tracks and 75 sessions, we’re diving deeper into the technologies, strategies and real-world lessons behind taking AI, ML and agentic systems from concept to production.
And this year, we’re bringing MLOps World right to the University of Texas at Austin campus at the Etter Harbin Alumni Centre, putting you in the middle of the action, and with Longhorn football season in full swing, there couldn’t be a better time to experience Austin. 🤘
Come for the AI, connect with the community, and enjoy a little of that Austin energy while you’re here.🤠
🎟️ Join us online November 16, then keep it going with us in Austin November 17–18.
Hope to see you there!
Good to know
Highlights
- 7 hours
- Online
Location
Online event
Agenda
-
Silent Drift: Why Your LLM's Quality Is Degrading and Your Metrics Can't See I
We caught our LLM quality problem the wrong way: through customer complaints. Three weeks of degraded responses had already shipped. Every monitoring signal we had was green the entire time because we were measuring the wrong things. Latency, error rate, embedding similarity: none of them captured what was actually happening, which was that response quality had quietly gotten worse in ways users noticed but our dashboards could not. After that incident we rebuilt how we think about LLM observability. The core insight was that output quality is not the same as system health, and you cannot infer one from the other. We needed a separate signal layer built from production behavior: how users responded to answers, where downstream tasks broke down, where retrieval and response stopped agreeing with each other. Getting this right took a few tries. A naive implementation fires constantly on normal LLM output variance. The real work was designing alerts that distinguish actual degradation fr
-
TBD
There's No One There to Press Retry: What We Learned Running AI on a Million D
"Everyone building on frontier models hits the same wall: 80% of the way there in a weekend, then an exponentially expensive climb toward the 99%+ that operational systems need. This talk is the honest map of that climb, from a team that now runs AI over millions of documents, images and videos a month for customers in logistics, fleet management, automotive and financial services who need the answer to be right every time. We will walk through the ladder nobody budgets for: retries, then queues when the model is down for three hours, then rate limits, then discovering that 95% is not enough for transactional data, then discovering that the model cannot tell you how confident it is. We will show two failures from our own production history, a customer whose single engineer racked up $30K of usage in a month because nothing was watching, and our first churn, on 500-page reports with a hundred rows per page, a problem we still consider unsolved. And we will show what we changed: a harne