I don’t know how tech savvy you are, but I’m assuming since your on lemmy it’s pretty good :)

The way we’ve solved this sort of problem in the office is by using the LLM’s JSON response, and a prompt that essentially keeps a set of JSON objects alongside the actual chat response.

In the DND example, this would be a set character sheets that get returned every response but only changed when the narrative changes them. More expensive, and needing a larger context window, but reasonably effective.

I have this battle - I am great at routine but terrible at habit. My wife asks me why I do the same thing every day, and I can’t really explain that I have to do it every day or i’ll stop doing it completely.

Listen. Well done. Just because it’s simple, doesn’t mean it’s easy.

Now go and put away your laundry. /s

I look after the strategy team for the CISO of a financial services company. I do enjoy my job, but I’ve swapped debugging IT systems for debugging organisational systems, and there’s a frustrating amount of baggage (financial reporting, process reporting, people managment) that you have to carry to get to that table.

I feel this. I used to have a job that involved popping difficult problems off the ‘hard problem’ queue, solving them and moving on. No one bothered me, I knew the set up and all was good. But it didn’t pay well, and now I manage people who get to have all the fun while I chase 30 minutes of focus in an 8 hour day.

I like to take it a step further - make the act of getting ready for the task a separate task. Other folks might see a single job, but when I have some repair work needing done around the house, I need a job to check if I have what I need to fix it, another to work out what I need to do, another to move it all to right place etc.