I really hate having to wait 20 seconds to a minute between every interaction with the LLM— I end up alternating between prompting and doom scrolling for several hours, a viciously unsatisfying cycle. (I know I could probably fix this by having multiple agents running at once, but context switching to that level also seems like a stressful doom-scroll-esque experience lol)
as someone who considered doing a phd but instead took the path of working for 4 years at Google, this was an interesting read :) software engineering and making a real world impact™ (plus money and perks) were fun for the first few years, but it's slowly gotten more monotonous and I've been missing thinking about / working on interesting problems. I told myself that in theory I should be able to just... do research on the side, but the 9 to 5 really gets you haha. certainly tradeoffs to taking the safe and economically rewarding path
hi all! first hackernews post and probably an underwhelming one — wanted to share this pirate bananagrams game I semi-vibe coded earlier last year :)
honestly it's way too easy to create stuff these days so I know it's nothing special, but it's my favorite board game and would love to find ppl to play with
reply