Stack Overflow

For 15 years, the Stack Overflow Developer Survey has tracked how the tech world actually works 🌐 This year, we want your honest take on everything from your favorite programming languages to whether autonomous agents are actually helping you code—or just making a mess. Have your voice heard in this year's #DeveloperSurvey results: https://lnkd.in/gDFRW-_6

Signature-based detection works exactly as designed, but attackers don't. Machine learning and autonomous agents are changing detection entirely, shifting from "does this match a known pattern?" to "does this actually make sense in context?" From The Heap, Samaresh Kumar Singh breaks down how SnortML and agentic AI work together to catch novel attacks in context. https://lnkd.in/gfDENk5Y

🎙️ We welcome Benny Yufei Chen, co-founder of Fireworks AI, to the show to explore what actually makes an AI application good or not, how to balance qualitative signals with quantitative metrics when evaluating AI, and how open-source eval protocols and community efforts are setting the standard for AI evaluation. https://lnkd.in/ghmwmnsx

🎙️ We sit down with Frank Portman, CTO at Yobi AI, to talk about why next-token prediction, is great for language but isn’t right for forecasting human behavior, how Yobi builds a “foundation model of behavior” using transformers, and what it takes to run millions of personalization decisions per second while keeping consumer data private. https://lnkd.in/gUxZKn2D

🎙️ In this episode of Leaders of Code recorded at Snowflake Summit, we're joined by Vivek Raghunathan, SVP of engineering at Snowflake, to chat about the five-stage framework his org used to go from "let chaos reign" to a repeatable, org-wide system for AI-assisted engineering; how Snowflake used "focus weeks" to push the frontier further; and how Snowflake cut release validation time from 15 days to a single day. https://lnkd.in/eY3cmwkJ

While agentic AI use is becoming more and more common in our modern software development lifecycle, developers in senior executive (50%) and software architect (52%) roles are by far the most likely to use AI agents on a daily basis. Academic researchers seem the most hesitant to adopt agents, with 22% not using AI-assisted tech at all. Explore more results from our latest #StackOverflowKnows survey: https://lnkd.in/eUny8YMi

🎙️ We sit down with Anish Agarwal, CEO of Traversal, to chat about why AI coding agents have made writing code easier but running it safely in production harder, what's really causing production failures (spoiler: it's not just code), and how teams can troubleshoot more effectively when traditional observability tools fall short with AI workflows. https://lnkd.in/egxfPtHJ

For the last decade and a half, the Stack Overflow Developer Survey has been a pulse check on the developer community 💻 This year, we want to know your thoughts on everything software, from your favorite programming languages to what it's really like to use autonomous agents. Learn more about our Annual #DeveloperSurvey and add your voice to this year results: https://lnkd.in/eUVdKBnh

When high-stakes agents—ones that can move around money or edit critical records—are built with human-in-the-loop review processes, human error caused failures are inevitable. But what if we built these agents around responsibilities, not capabilities? In this Dispatch from O'Reilly, Artur Huk explores the Responsibility-Oriented Agent framework and how it could be the solution to our alert fatigue. https://lnkd.in/e4gdKQtu

It's probably no surprise that daily AI agent use has skyrocketed since our 2025 #DeveloperSurvey. According to our May survey on agentic AI in the workplace, more than a third of developers are using autonomous AI tools everyday in their work. But the number of respondents who don't use or plan to use AI has also increased, now accounting for 21% of developers. Explore more results from our latest #StackOverflowKnows survey: https://lnkd.in/exXNbKSp