Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore how to evaluate AI investment quality and deployability for measuring mission success in this 47-minute conference talk from Databricks' Data+AI World Tour for Government 2025. Learn to move beyond basic reporting by quantitatively measuring operational AI outcomes and linking them to high-level strategic goals. Discover Databricks' latest science for quantifying AI quality and see how unified data environments enable continuous feedback loops that monitor both program performance and mission objectives for constant improvement. Master practical approaches for establishing reliable metrics for subjective evaluations, techniques for calibrating LLM judges to enable cost-effective and scalable assessment, and actionable frameworks for evaluation systems that evolve with AI capabilities. The session covers why measurement is the real bottleneck in AI, must-haves versus must-not-haves in AI systems, measuring AI with calibrated LLM judges, and includes a real-world case study on measuring mission impact at the World Bank Group presented by their Head of Data & Analytics.
