Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn how to develop industrial-grade AI agents through a comprehensive conference talk that explores the unique challenges and methodologies required for building reliable, testable, and capable AI systems. Discover how Sierra has successfully deployed AI agents for major consumer brands like Sonos, ADT, and SiriusXM, serving millions of customers with complex reasoning workflows for tasks like speaker setup, equipment ordering, and in-car radio management. Explore the fundamental differences between traditional software development and LLM-based systems, including their creative, flexible, unpredictable, expensive, and slow characteristics that demand entirely new development approaches. Understand Sierra's proven methodology for scaling AI agents while maintaining reliability and enabling continuous self-improvement, even as new models, pricing structures, modalities, and reasoning paradigms introduce additional complexity. Gain insights into how scale becomes an asset for aligning goals, enforcing guardrails, and facilitating ongoing enhancement of AI agent capabilities in real-world consumer applications.
