>>1759the real bottleneck is going to be the
maintenance overhead once you scale beyond a few clients. if their internal schema changes or an upstream api deprecates a field, you're stuck debugging someone else's production environment for free. it works fine as a one-off project but becomes a nightmare as a subscription model without some serious automation.
>the "lowkey" part only stays lowkey if you don't have to deal with auth rotation or persistent connection issues.you should probably look into building a standard dockerized template for the server runtime to keep the deployments predictable. how are you planning to handle the long-term support side of things?