Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore how MLflow is evolving beyond traditional machine learning to support generative AI, agents, and production-grade systems in this 58-minute podcast episode featuring three Databricks experts. Discover why MLflow isn't just for data scientists anymore and learn how the platform is being rebuilt to handle the complexities of modern AI workflows, including messy evaluations, memory management risks, and governance requirements. Gain insights into MLflow's transformation from Corey Zumar (Product Manager focusing on MLflow and LLM evaluation), Jules Damji (Lead Developer Advocate working on Spark and lakehouse technologies), and Danny Chiao (Engineering Leader specializing in data and AI observability). Understand the critical shift from treating agents like simple chatbots to implementing proper production systems, examine AI UX design patterns and context management strategies, and learn about human feedback integration, prompt optimization techniques, and the expanding personas using MLflow. Delve into discussions about product ecosystem design, the balance between persona expansion versus separation, and the crucial distinctions between PII and business sensitivity in AI systems. Master the evolution of MLflow as it adapts to support the full spectrum of modern AI development, from traditional ML workflows to sophisticated agent systems requiring robust governance and observability.
