drdrey·ieri·discussthe problems are general and abstract, domain specific knowledge and memorization don't help. Figuring out the rules, the goal, the controls, and how to solve in a reasonable budget all indicate some level of general ability.
drdrey·mese scorso·discuss> And it can pay for itself twice: once in real work shipped, and once again in something else you could probably use more of, which is gravity.what?
drdrey·mese scorso·discussas someone who doesn't know how to get better design out of LLMs, can you elaborate?
drdrey·2 mesi fa·discussI have finally embraced squashing PRs and realized I wasted my youth trying to write Good Commits.
drdrey·6 mesi fa·discussthis seems like what MCPs should have been from the beginning. If you think MCPs are better, can you explain why?
drdrey·6 mesi fa·discussthe problem with any approach like this based on usage metrics is that it will be abused to death
drdrey·7 mesi fa·discuss5.2-codex just came out. You could use codex with regular 5.2 for a week or so.
drdrey·7 mesi fa·discussok but these are details, the point is that the operators of the database are external, selfish and fluctuating
drdrey·7 mesi fa·discussthe big difference is the trust assumption, anyone can join or leave the network of nodes at any time
drdrey·9 mesi fa·discuss> you don’t have to follow the herdmeanwhile, OP's entire substack post denouncing hustle culture is about plugging his viral tweet and his ebook
drdrey·9 mesi fa·discussit's hard to take the article seriously when the author relies on a fuckgit alias on a regular basis
[ my public key: https://keybase.io/dreynaud; my proof: https://keybase.io/dreynaud/sigs/ClYpOl4mWSBHYzAwHMzscXuzBJffIQ0D9gqygBHQTlg ]