Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore comprehensive guardrail technologies and strategies for building trustworthy AI systems in this 19-minute intermediate-level conference talk from Databricks. Learn how Northwestern Mutual leverages various methods within the Databricks platform to deliver and evaluate guardrail models for AI safety, covering techniques from prompt engineering with custom-built frameworks to hosting marketplace-served models. Discover how to utilize GPU clusters for fine-tuning and running large open-source models at inference, including practical implementation of Llama Guard 3.1, and understand approaches for generating synthetic datasets based on production questions. Gain insights into real-world AI governance practices and safety measures that can be applied to enterprise AI deployments, presented by Nicholas Brathwaite from Northwestern Mutual's Data Science team.
