Hacking with LLMs

Over the last few months I’ve been spending a fair amount of time programming and doing so with LLMs is different. Probably better. At least…for some kind of definition of better. It allows me to achieve much more in a much less time which sounds like it’d be net positive. Simple idle questions can easily be solved without much investment. So, why don’t I just outright say that it’s better? Because I feel like you lose something through LLM-based programming. I’ve produced something, but do I really understand it? Often not. Let me give you an example of a project I recently worked on… Morrisons More My wife and I tend to shop at the Morrisons in Boroughbridge and they have a loyalty card program (who doesn’t)? When poking through that online account I found that they have digital receipts, showing everything that I’ve bought at Morrisons over the last year. It’s all rendered through a painful Javascript system so no raw HTML and each Javascript call is protected with multiple…