i have a rule, since working with jetbrains junie llm codining agent
if it spends an hour, trying three times to do something, i probably can do it in 10 minutes. already have had this happen twice. it seems to be turning into an operational policy.
i have a rule, since working with jetbrains junie llm codining agent
if it spends an hour, trying three times to do something, i probably can do it in 10 minutes. already have had this happen twice. it seems to be turning into an operational policy.
to explain this, the logs of junie often help me figure out how to solve the problem, and then i get clarity about how to to actually write the code, and that's the reason why i have succeeded at quickly, because the AI tried a dozen ways to not succeed