I feel like I’m missing something between what is explained here and claims that agent driven coding massively speeds up output. Certain parts of this like Day 2s decomposition into steps and Day 3s context packet seem to imply that the author already knows the codebase. They know what steps need to be taken to fix the problem and they know what abstractions exist. That all makes sense to me and reflects what I’ve experienced. It helps get done what I already know. But I still have to understand and know the codebase which feels at odds with some of the massive upside some people are selling. This might be a little adjacent to the main point, but overall I’ve had similar good experiences with the points raised on day 2 + 3.
Back on topic though, I wonder how much can be gleaned from reading reddit (and twitter) about such a narrative-driven topic. At this point, I figure that everyone is either lying, not a person, insane, or trying to sell something. As soon as I stop reading from these platforms and their messages, the narratives they sell cease to exist, and things become much more normal. There was a time when I felt like online discussion was helpful in learning and expanding horizons, but now it feels like it narrows my experience and makes it more myopic.