Building Agentic AI Workflows Using Fusion MCP Server

This week we welcome back Senior Technical Evangelist Mike Nelson and introduce podcast newcomer and Product Manager Ameerul Shah. Our conversation centers on the latest developments with Fusion and how it is redefining and automating data management. We explore how the Everpure is moving beyond simple storage management by leveraging the Fusion automation engine to empower IT teams to become true data strategists. A major focus of our discussion moves to the new Fusion MCP server release. Mike and Ameerul explain the Model Context Protocol, describing it as the important bridge between infrastructure data and large language models. By providing the right context, the new Fusion MCP allows AI agents to move past confident guessing and hallucinations, enabling them to safely execute real world tasks like provisioning workloads and resolving incidents with supervised actions. We then highlight the open source commitment behind Fusion MCP, designed to help power users and dark site customers integrate their full infrastructure stacks. Mike and Ameerul discuss the power of API first development, the importance of human in the loop supervision for agentic AI, and the broader shift toward autonomous management. Listeners will gain practical insights into how this technology is helping organizations streamline operations and scale their data services more effectively. To learn more, visit: https://blog.everpuredata.com/purely-technical/smarter-ai-powered-data-management-with-everpure-fusion-mcp-server-2/ Check out the new Everpure digital customer community to join the conversation with peers and Everpure experts: https://purecommunity.purestorage.com/ 00:00 Intro and Welcome 06:38 What Product Managers Actually Do 09:55 Origin of Fusion 14:05 Applying AI to Data Management 16:18 What is Model Context Protocol (MCP) 25:45 Observe, Triage, and Act 35:15 MCP is Open Source 38:25 Availability of MCP for Users 42:49 Hot Takes Segment 48:31 Oops IT Stories