# How Zillow maintains context in real estate processes with AI At VB Transform 2026, Zillow's AI architecture and context management were discussed. - **Source:** [VentureBeat (feeds.feedburner.com)](https://brifing.si/source/c51474dc-6b09-49dd-aaca-085f4ed00c84) - **Published:** 2026-07-20 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/1d2ff22f-d167-4d76-8135-53b0944132eb ## Zillow's customer journey Zillow tracks the complex journey of customers from phone screens to loan officers to real estate agents. Maintaining context throughout this process is impossible with a single chatbot. ## The importance of the context layer While Zillow's AI efforts started with data, the real challenge was remembering and carrying forward the context of the customer's journey. This context layer is crucial for understanding the customer's stage. ## Custom AI architecture Instead of routing customers through a single model API, Zillow developed its own AI architecture. Leveraging 20 years of machine learning experience with products like Zestimate, they created task-specific models. ## Glean integration Zillow automated thousands of repetitive tasks with Glean. Glean's MCP gateway centralized connections for finance, legal, and marketing teams, reducing costs. ## AI cost optimization Glean's model routing and precomputed context mechanisms reduce token consumption, controlling AI spending. This focuses on preventing waste rather than adding new capabilities. ## Lessons for enterprise AI Zillow and Glean's approach highlights the importance of establishing a measurement baseline upfront, centralizing context, and layering permission management in enterprise AI projects. ## Sources covering this story - https://venturebeat.com/data/at-vb-transform-2026-zillows-engineering-chief-said-ai-roi-numbers-only-hold-up-if-you-measure-before-you-build — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._