-
I kept hitting Claude Code rate limits way earlier than made sense. Not because I was doing anything at scale. I wasn’t running massive jobs or processing huge datasets. I was just using it โ normally โ and somehow burning through tokens faster than I expected. Turns out most of the inefficiency wasn’t from what
-
The biggest learning as a PM transitioning into AI: “๐ช๐ผ๐ฟ๐ธ๐ถ๐ป๐ด” ๐ถ๐ ๐ป๐ผ ๐น๐ผ๐ป๐ด๐ฒ๐ฟ ๐ฏ๐ถ๐ป๐ฎ๐ฟ๐. In traditional software, PRDs define requirements, engineering writes test cases, and you validate against clear pass/fail criteria. That works when “does it work?” has a yes or no answer. AI doesnโt behave that way. It can pass predefined test cases and
-
The $28 billion business model hiding inside your boarding pass โ and what it means if you’re building in this space. Here’s a fact that took me a full minute to process. During COVID, when Delta Air Lines needed emergency cash, they didn’t pledge their planes. Not their gates. Not their routes. Not their slots
-
And learned the most important lesson about vibe coding along the way. I believe in learning at least a little bit of the local language of wherever I live – even if it means suffering through embarrassingly broken conversations. To my surprise, people have usually been super supportive of my attempts rather than laughing at
-
Ten days into my job search, I was spending hours scrolling and still applying to the same eight roles. The problem wasnโt effort. It was signal. So I built a small AI tool to handle discovery. It scans multiple platforms, filters and deduplicates listings, and surfaces only roles worth my time. In one run, 127 listings became 43 real options. But I deliberately broke the Apply button. No auto-apply. No auto-fill. Because a job application isnโt a workflow. Itโs a choice. I automated the noise and kept the decisions human.