Abstract: 

The largest issue ML teams face is due to a gradual decline in model performance that is hard to catch in time. It’s always in hindsight that they realize the cost of inaction of an underperforming system.

On average, most generative models are truthful only 25% of the time, according to Stanford Artificial Intelligence Index Report 2022 . The downstream impact is not just limited to ML or business teams as it trickles down to the end customer who faces the first-hand repercussions of inaccurate and incorrect predictions. With the adoption of AI Observability tools, pioneering tech giants have managed to quickly catch model issues, diagnose root causes, and perform continuous improvement to optimize the model’s performance.

Join a session with Ayush Patel, Founder & CEO at Censius as he discusses the current state of AI/ML, the top reasons for silent model failures, and how they can end up costing a lot of time, resources, and dollars. Explore strategies to proactively gauge performance dips and re-train
models and build reliable and compliant AI solutions.

Key takeaways
● The state of AI in the real world and its impact
● Understand the ‘why’ behind silently failing ML models
● Best practices to proactively spot and fix model failures to avoid costly mistakes

Bio: 

Ayush is the co-founder of TwelveFold, an AI start-up studio, where he manages a portfolio of MLOps and Generative AI companies with entrepreneurs. He also works as the CEO of Censius, an AI Observability platform that helps to optimize AI models' real-world performance. As a seasoned professional, he has closely worked with customers across industry verticals, AI teams, and research projects to build reliable and compliant AI solutions to solve everyday business problems and scale models at production.