This is correct, find a problem worth solving first and ideally know your first customer before you even start. The pitfall that I as a transitioning SW eng fel

This is correct, find a problem worth solving first and ideally know your first customer before you even start. The pitfall that I as a transitioning SW eng fell into was to build the most challenging and complex solution I could and then find a use case. Don't do this, find a problem and solve it with the simplest solution you can.