Satish Hegde

RAG got you the demo. Context engineering gets you from pilot to production. The gap between a working pilot and a production-grade agentic system isn't better prompts or bigger models - it's the data and knowledge infrastructure underneath. Production agents need more than document retrieval: they need governed data products, semantic resolution, multi-source federation, domain knowledge graphs, versioning, and provenance - all delivered through a unified context layer. In this session, we present the Unified Context Stack built on top of lakehouse platforms based on our experience from enterprise deployments. We walk through the building blocks that production agents demand but pilots ignore: a Context Graph Fabric, Unified Access Layer and an Agents plane powered by Databricks ecosystem. Join us to learn our phased adoption journey - starting with metrics view on your existing lakehouse and evolving incrementally toward an agent-ready context platform.

Toyota, the world’s largest automaker, sought to accelerate time-to-data and empower business users with secure data collaboration for faster insights. Partnering with Cognizant, they established a Unified Data Lake, integrating SOX principles, Databricks Unity Catalog to ensure compliance and security. Additionally, they developed a Data Scanner solution to automatically detect non-sensitive data and accelerate data ingestion. Join this dynamic session to discover how they achieved it.

Today, field service operations often face challenges stemming from disconnected data, limited real-time information, and difficulty accessing the right expertise at the point of service. Technicians may struggle to diagnose complex issues, leading to repeat visits, extended downtime and frustrated customers. This session will delve into how we can move beyond these limitations by creating a truly "connected technician" experience, leveraging the power of agentic AI within the unified data environment of Snowflake.