Session: Model Context Protocol for Humans: Applying AI System Design to Team Leadership
Every AI model has a protocol, a system that delivers the right context, tools, and mission so it can perform at its best. When did we last design the same for our teams?
MCP(Model Context Protocol) is the open standard that connects an AI model to the real-world data and tools it needs. In this talk, I'll show how the same architecture applies to engineering leadership. Your team is the model. You are the MCP Server. And your job is not to direct, it's to connect.
Drawing on real examples from Starbucks, we'll explore how three primitives- Resources, Tools, and Prompts to eliminate bottlenecks, give teams the context they need to self-direct, and create environments where performance emerges naturally. If your protocol is right, collaboration doesn't have to be managed. It emerges.
Bio
I’m Leela Kumili, Lead Engineer at Starbucks specializing in distributed systems, microservices, and cloud architecture. I write the LinkedIn newsletter The Engineer’s Path about engineering leadership, AI adoption, and career growth, and I also serve as a news editor for InfoQ, covering architecture and distributed systems. My work helps engineers bridge technical excellence with system-level leadership.