Choose Your Stack

The world of software has changed more in the last year than the previous twenty. AI is the reason why. What does that mean for the people building in it? From agent permissions to software craft to what the next 16 years could actually look like, this episode gets into the practical realities of building in the agentic era from someone who has been doing it since 1999. In the fifth episode of Choose Your Stack, recorded at the MCP Dev Summit in New York City, Ryo Koyama sits down with Jamu Kakar, co-founder of Valet, a cloud built for agents that lets enterprises create and run agents defined entirely by prompts. Jamu breaks down why giving agents a CLI creates real security risks, why the craft of software development isn’t dying but transforming, and why despite his moments of doomerism, he believes AI is opening up a range of creative experiences for humans that we can’t yet imagine. 0:35 What is Valet? 1:55 MCP vs CLI: How Fast Is the Agentic Era Moving? 4:28 Anthropic Skills and MCP: Finding the Right Primitives 6:01 Better Models vs Better Prompting: What Actually Drives AI Productivity? 8:07 Is Cloud Infrastructure Ready for AI Workloads? 10:05 How to Get Real Efficiency Gains from AI 12:03 Why Human Connection Still Matters in an AI World 12:43 Growing Up Without Knowing a World Before AI 15:02 Hot Take: AI Will Unlock a New Era of Human Creativity Are you building for the world that exists today or the one coming next?

Would you invest in a company with no humans? From AI economics to agent mortality to the generational divide over what humanity actually means, this conversation goes places most AI podcasts are too cautious to go. In the fourth episode of Choose Your Stack, recorded at the MCP Dev Summit in New York City, Ryo Koyama sits down with Emma Seymour, investor at Arthur Ventures, a growth equity firm backing the fastest moving founders in AI and machine learning, to explore what the agentic era looks like from the other side of the cap table. Emma breaks down why the companies generating the most economic value are the ones using AI to complete entire units of work rather than just assist humans, what happens to a world dependent on agents when those agents disappear, and why AI is going to teach us more about humanity than anyone expects. 0:00 Intro 0:16 What Emma does 1:02 AI hot take: every agent will die 1:50 What happens when AI becomes infrastructure and then disappears 3:15 AI governance, governments, and propaganda 3:53 What companies will generate the most economic value 6:24 Would you invest in a non-human company 8:04 Generational adoption of AI 10:21 Wall-E, the Matrix, and what AI means for humanity Are we ready for what comes next? Or are we closer to Wall-E than anyone wants to admit?

What happens when the people who could never fully use computers finally can? From open source strategy to agentic interfaces and developer tooling, almost every assumption about who gets to build with technology is being rewritten. In the third episode of Choose Your Stack, recorded at the MCP Dev Summit in New York City, Ryo Koyama sits down with Chris DiBona, Vice President from the Office of the CTO at Microsoft, to explore what decades of open source taught us about developer ecosystems and why those lessons matter more than ever in the age of AI agents. Chris breaks down why MCP succeeded where other protocols failed, why the line between programmer and non-programmer is disappearing, and why the real challenge for the industry is building AI tools that are personalizable and secure for billions of people who do not all look or think the same. 0:00 Intro 0:21 Microsoft’s open source contributions 2:07 The excitement behind MCP 5:22 CLI vs MCP 8:04 The democratization of coding 11:23 Chris’s Hong Kong cabinet AI project 14:04 Agents as toil reducers 15:19 The future of developer relations 17:04 Open source and LLM training 19:40 Hot take: the industry doesn’t know real programmers 21:08 Personalization at a billion-user scale The computer industry has been making the same promise for decades. Are AI agents finally the ones to keep it?

What happens to human work when AI agents can do most of it? From agent governance to model personality to generational adoption, the conversation around artificial intelligence is no longer just about what agents can do. It is about what humans become in a world where they manage them. In the second episode of Choose Your Stack, recorded at the MCP Dev Summit in New York City, Ryo Koyama sits down with Jake Mannix, Technical Fellow, AI, Relevance, and Agentic Governance at Walmart, the world’s largest private employer, to explore what the agentic era actually looks like at enterprise scale. Jake breaks down why the future of work is not humans versus agents but humans learning to manage them, why enterprises will use many specialized models rather than one universal AI, and whether large language models actually have an inner life and what that means for how we build with them. 0:00 Intro 0:20 Jake’s role at Walmart 2:33 Should you treat AI agents like they have feelings? 4:27 How to manage agents effectively 5:04 LLM Culture wars 6:22 Why enterprises need many specialized models not one universal AI 8:20 How fast are open source models closing the gap? 10:14 What skills will matter most in the AI era? 13:36 Do LLMs actually have a first-person experience? 14:25 Is AI bigger than the internet?

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?