miniblog.

LLMs have been really helpful for me porting some JS projects to TS. Since the types are erased and annotations are optional, I can go gradually and safely with more minimal human oversight. (I'm not sure if it'll significantly help robustness, but it's lovely for navigation.)

After further playing with my LLM project, I'm surprised how hard it is to tune with system prompts. My agent kept saying "obviously" even though my prompt said "helpful, professional". Eventually I found that a "courteous" prompt gets "of course" which is better but not ideal.

I regularly see the phrase "all Xs are Ys, but not all Ys are Xs" in teaching material. Even material for children! I have to re-read it every time. I very much prefer "Y is a more general category than X" or "X is a subset of Y". Do people find this phrasing helpful, or is it poor pedagogy?