Stack Overflow

There's one question we can no longer avoid with autonomous AI systems: Do we really need to govern every decision? In this Friday column from O'Reilly Radar, Varun Raj explores finding the balance between autonomy and governance in AI production systems, and how you can use selective control to get the safest and most useful results out of your AI tools. https://lnkd.in/exJPswGn

🎙️ We welcome Harsha Chintalapani, co-founder and CTO at Collate and co-creator of Open Metadata, to the show to discuss why AI and LLMs struggle with real-time production data, how inconsistent definitions and weak governance can break both your analytics and MLs, and what companies can do to get their data AI-ready. https://lnkd.in/ecdQS3WK

AI agents are only getting more powerful, with protocols like MCP making them priceless tools for the everyday worker. Why wouldn't the Worst Coder in the World want to build one? Phoebe Sajor, our self-proclaimed "coding ingénue," discovers the challenges and rewards of building an agent to crack the Stack Internal leaderboard, and learns a few things about the beauty of coding along the way. https://lnkd.in/d7FJiSre

Turn your organizational noise into high-signal, verified knowledge with Stack Internal! 🚀 Ingestion harnesses scattered content across your teams and tools and transforms it into structured, trusted intelligence that powers AI initiatives and agentic workflows. Learn more about Ingestion, now generally available to all Stack Internal Enterprise customers: https://lnkd.in/eiynewY8

🎙️ We're joined by Cult.Repo's Emma Tracey and Josiah McGarvie to discuss making documentaries about open-source software, why open-source projects and the people who maintain them are such interesting stories for audiences, and what they see as the common stressors that plague all open-source projects, like sustainability and burnout. https://lnkd.in/eVKQ3sG7

We're no longer in the "mess around" stage of AI—technologists and organizations alike are using AI tools to build products and solve real-world problems. From autonomous agents and enterprise monetization to anxieties about AI's cultural and economic impact, Ryan Donovan reports from the floor of #HumanX2026 what it means to be in the "find out" stage of AI. https://lnkd.in/eWV-E95g

🎙️ Mike Swift, co-founder and CEO of Major League Hacking, joins the show to chat about the never-ending need for developer communities, how MHL’s recent acquisition of DEV brings together building and knowledge sharing, and why now is the best time to be both an artisan and a builder of software. https://lnkd.in/eEySHc95

Adopting AI is no easy feat, especially in large organizations. During the AI era, one thing has become clear from our work with enterprises—the quality of your data is more important than you think. And the foundation you build of human-validated knowledge can make or break your AI strategy. Build your knowledge foundation for AI success with Stack Internal, our gold-standard context layer: https://lnkd.in/ehN26beT

AI can feel like a black box: prompts go in and outputs come out. But the code they produce is often a tangled mess of incorrect assumptions and inexplicable choices. Jonathan Gordon, CEO and co-founder of ReWeaver AI, explores what causes black box AI drift and why the solution isn't simply keeping "humans-in-the-loop"—its giving humans back the control. https://lnkd.in/eBigQJuk

#SPONSORED 🎙️ We welcome Intuit's Chase Roossin and Steven Kulesza to the show to chat about the hardest problem in engineering right now: getting multiple AI agents to work together in a complex system, plus how automated evals can make agents more predictable and how their customers shaped their technical architecture. https://lnkd.in/eJqp3DNC