Technical Talks

Han-chung Lee

Machine Learning Director | Moody's Analytics

The Model is the Product

Foundation Models

In the realm of machine learning, AI, and deep learning, the intelligence embedded within a system—the model—stands as the primary product and key differentiator. This talk explores how the intelligence component has evolved to become the central selling point across technological eras. We will examine the historical progression of how intelligence capabilities have increasingly defined product value, transforming from hardware differentiators like "Intel Inside" during the PC era, to software advantages, and now to model-centric offerings in today's AI landscape. The intelligence layer has become not just a feature but the core product itself. Additionally, we'll analyze how the definition of "model" itself has evolved alongside technological advancement, reshaping what constitutes a system's core value. Companies now face a strategic bifurcation: pursue a model-centric approach or focus on distribution-centered strategies. Each path carries distinct trade-offs, risks, and opportunities in today's competitive AI marketplace. Through case studies of industry leaders and emerging players, we'll demonstrate how the fundamental principle—"the model is the product, the distribution is the moat"—is reshaping competitive dynamics and business strategies across sectors.

Machine Learning Director

Han-chung Lee

Moody's Analytics

Han-chung Lee is a machine learning expert who builds and operates AI systems with a focus on GenAI, LLM agents, and recommendation engines. Currently Director of Machine Learning at Moody's Analytics and founder of Calabazas Creek, Han-chung excels at untangling complex code and organizations. A Berkeley EECS grad with an MBA from SJSU, he shares insights on ML engineering and tech investing drawn from his diverse experience across sell-side, buy-side, and venture capital. Follow his practical wisdom and occasional industry reflections.