In response to the surge in demand during the COVID-19 pandemic, Instacart rapidly expanded its operations, partnering with over 65,000 stores and serving more than 85% of U.S. households, necessitating a significant overhaul of its data infrastructure. The company transitioned from batch to stream processing using Kafka and Confluent Cloud to manage the vast data generated by around 10 million daily customers, ensuring data quality and discoverability while maintaining a reliable IT architecture. This shift facilitated the creation of a Customer 360 view and improved inventory management, allowing Instacart to integrate real-time data into its systems and enhance predictive models for better customer service. By adopting a data mesh strategy, Instacart clarified data ownership and standardized processes, empowering teams to access and utilize high-quality data efficiently, ultimately improving algorithms and personalizing customer experiences. Confluent Cloud's serverless capabilities and infinite storage have become central to Instacart's data strategy, streamlining operations and allowing the company to focus on expanding its offerings without worrying about infrastructure constraints.