1. 14

  2. 1

    Congrats! This is a cool writeup….these posts always motivate me to be more productive

    1. 1

      On the final note about building for simplicity versus scale, I think this project was fortunate. 2000 requests per day is pretty manageable. Even 2000 requests per day, as the author points out. Based on the graph, these were the two largest leaps of traffic to this service. Aside from that, the growth is very linear and predictable, which affords for good planning and resource management. I’ve never had a successful project like this but I remember reading lots of stories about how people scaled up and most of the rush is when this scale up is unpredicted and extremely sudden, like that one web app where you talk to an AI.

      I find it interesting how clearly defined the weekday traffic versus the weekend traffic is once you crossed the 500,000 threshold. I wonder if it’s just scale, or if there is statistical significance there.