How Many CTOs · 2026-06-02 · 33 min
In this episode of "How Many CTOs Does It Take?" podcast, hosts Scott Porad and Brad Hefta-Gaub discuss a high-performing engineering team that has effectively cleared its product backlog while using AI heavily, including a senior engineer who surprised Brad by running agents overnight to automatically perform fast, high-quality GitHub PR code reviews via a webhook-driven harness. The engineer built an opinionated, RAG-backed "world-class code review" skill set that pulls new PRs from selected repos, reviews them with defined criteria, and posts comments, highlighting the need to articulate engineering judgment rather than relying on gut feel. They explore how models and agent frameworks are becoming commoditized, making the key differentiator the ability to assemble components and codify professional, reusable skills. The conversation turns to how teams should integrate AI into the SDLC, balancing experimentation with a shared, periodically updated "long-term support" approach and quality gates, plus examples like agents generating test coverage and proposing fixes for failing tests.
Introducing Skein and Beaker Stack: Using AI Agent Engineers to Ship SaaS Faster67 / 100AI and the Data Swamp In Your Backyard with Ben WilcoxPioneering Music Education with John von SeggernRemoving Bottlenecks, Not People with Stephanie SylvestreNavigating the Future: Gas Town and Bespoke AI Agents