🧠It looks like LLMs are making devs 10x more productive.<br /><br />But something else might be... Over the past few months, I’ve been watching a pattern emerge.<br />Devs rave about LLMs massively boosting their output. PMs boast about faster features. Founders post screenshots. VCs… well, they think they found the next gold mine. But when I zoomed in, something felt off.<br /><br />The devs getting the most value from AI agents aren’t doing anything “new.” They’re doing this instead:<br />✅ Writing tight, testable code<br />✅ Working in small, iterative steps <br />✅ Focusing on feedback and flow<br />✅ Getting clarity before coding<br />✅ Refactoring relentlessly<br /><br />Sound familiar? It should. That’s the original Extreme Programming (XP) playbook. And it’s looking more relevant now than ever—just wrapped in AI.<br /><br />🚀 So what’s really driving productivity here?<br />→ Is it the machine, or the mindset?<br />→ The assistant, or the agile habits? I<br />
