Should You Still Build Software for Humans? | Michael Grinich, CEO of WorkOS

What does it actually mean to build software in a world where AI agents are becoming the primary users? From open source strategy to agent identity and agent permissions, almost every assumption about how software gets built is being challenged. In the first episode of Choose Your Stack, recorded at the MCP Dev Summit in New York City, Ryo Koyama sits down with Michael Grinich, CEO and founder of WorkOS, the AI authentication layer behind OpenAI, Anthropic, Perplexity, and Cursor, to unpack what the shift to agentic software development actually looks like in practice. Michael breaks down why agent identity is the wrong problem, how traditional software permissions break down in an agentic world, and what every developer and founder needs to understand about building AI tools in 2026. 0:00 Intro 0:16 What is WorkOS 1:15 What has the AI transition looked like from your seat over the past 2 years? 3:21 Closed source vs open source and training AI Agents 4:51 Building for agents first, humans second 7:53 Is auth moving deeper into the stack? 11:46 AI hot take: What is everyone getting wrong about agent identity? The rules of software are changing faster than most companies realize. Is your product ready to evolve for a world it wasn’t designed for?