

Sometimes to the point of causing a live incident. Unfortunately, sometimes the feature did not actually work, or would not handle edge or error conditions which caused the program to be unstable. This insulted them, because there was no problem and solving approach, just bruteforcing code with live reload until the feature worked. In the face of that, the only thing I could do is: give them the benefit of the doubt by asking them to walk me through their problem solving approach. The reason: programming by trial and error. They would submit a lot of pull requests, but upon review it became very apparent that documentation was not being consulted resulting in unnecessarily hacky solutions. I worked with a large group of a local graduates from a well-known coding camp.
