I have the same experience. I'm in gamesdev and we've been encouraged to test out LLM tooling. Most of us at/above the senior level report the same experience: it sucks, it doesn't grasp the broader context of the systems that these problems exist inside of, even when you prompt it as best as you can, and it makes a lot of wild assed, incorrect assumptions about what it doesn't know and which are often hard to detect.
But it's also utterly failed to handle mundane tasks, like porting legacy code from one language and ecosystem to another, which is frankly surprising to me because I'd have assumed it would be perfectly suited for that task.
In my experience, AI for coding is having a rather stupid very junior dev at your beck and call but who can produce the results instantly. It's just often very mediocre and getting it fixed often takes longer than writing it on your own.
My experience is that it varies a lot by model, dev, and field — I've seen juniors (and indeed people with a decade of experience) keeping thousands of lines of unused code around for reference, or not understanding how optionals work, or leaving the FAQ full of placeholder values in English when the app is only on the German market, and so on.
But the worst LLMs? One of my personal tests is "write Tetris as a web app", and the worst local LLM I've tried, started bad and then half way through switched to "write a toy ML project in python".
The reward for those things is either more work for no increase in pay or you're in conflict with one or more people who are now bad mouthing you to the boss and you'll eventually be fired or be made so miserable that you'll voluntarily leave.
> She, on the other hand, was completely relaxed -- she spent her days working with Nobel prize winners and loads of other people for whom she had no doubt were smarter than her. Being confronted with loads of people smarter than her was a daily experience.
I'm a college dropout who has managed to work professionally, in software, in RF / embedded development, medical robotics, all over the web, and most recently in AAA games. I've called PhDs colleagues, people with multiple Masters, people with decades of industry experience at the top of their fields...
I always feel like the odd-man out, and, while it used to bother me a bit, I'm pretty content with the fact that I probably always will. I frequently feel like I'm inferior to my colleagues, because of the sheer depth of their knowledge as it relates to whatever the particular domain is.
But I have the sense that I'm doing something right because I get great reviews, I frequently find moments where I can teach my colleagues something they didn't know, and they come to me for help, advice, and say good things about me (and vice versa).
But it is still a very odd feeling and I think it'll be with me for however long I work in this industry.
The company was not very well run so I’m not surprised. Their stock price has tanked over 10x since IPO, and it dropped by half in the employee lockout period after IPO.
In a current PF2E game, my Cleric has a deity and I do RP him to stay in Ragathiel's favor. It's explicitly called out and I don't think it's dumb at all...
Sure. But is Ragathiel any better than Bjorn’er, the god of rapturous dance that I just made up? Imo, no. If someone wants to choose a predefined god, sure. If someone wants to make one up? Also sure.
The only thing I’d be fairly vocal about is that until some lore has reason to enter the narrative, it isn’t canon. E.g. the space faring races that appear in both dnd and pathfinder
I mean, objectively yes I do think Ragathiel is better than Bjorn’er because there is actual lore, thought, and consistency there. [0]
Look by all means, if you want to bring your own deity or $WHATEVER to a table I don't think most reasonable DMs and players would even bat an eye but you'll absolutely be expected to put some degree of effort into this beyond just showing up unprepared and cooking shit up on the fly.
I would argue it’s even better role play if players don’t know things though. If you meet some followers of Bane, a player who knows the lore will probably deem them to be a bad guy. If they have no idea, they will roll knowledge to see if their character would know. The DM providing information based on character knowledge checks is generally a great source of fun.
The DM saying “yeah I know you know Bane is a bad guy but I made your roll for it and you failed so you need to pretend you don’t know that” is never very good even if the players try to obey the spirit of things.
I'm fine today but the decade following high school graduation was really rough for me. Stints of homelessness, substance abuse, depression, run-ins with law enforcement (some deserved, some not). I ended up enlisted in the military to escape my situation and it worked out for me. I'm in AAA games today but have had a good career in software engineering and we could probably retire today if we were willing to move to a LCOL area. But we aren't, because...
Today we have a rich and deep group of close friends, mostly formed around our children, and I pour myself into my family obligations. I'm trying to develop small communities where I live and it's starting to pan out. Again, it's primarily based around our children, but we've developed some lasting connections with a couple of other parents.
I'm currently building a 2D 16-bit platformer with my (twin) daughters, who are 5. I let them pick out some asset packs to use and, once I get this level editor finished, will have them help build levels. My aim is to get it running on a SteamDeck (will just copy the required files over) and let them play it on the TV via the dock. They've seen a bit of Mario and Zelda so far and are curious for more. I figured this way would be a bit more hands-on and they could get some sense of how the sausage gets made.
> The key to making things work is having a cohort of parents that have similar priorities. If the parents in your social group default to shutting Junior up with an iPad, you're going to have a bad time.
This has been our priority as parents forming peer groups with other parents. But it's very hard to find the kids that your kids like and are friends with who aren't constantly inundated with tech.
I used the Copilot trial. I found myself waiting to see what it would come up with, analyzing it, and most often time throwing it away for my own implementation. I quickly realized how much of a waste of time it was. I did find use for it in writing unit tests and especially table-driven testing boilerplate but that's not enough to maintain a paid subscription.
But it's also utterly failed to handle mundane tasks, like porting legacy code from one language and ecosystem to another, which is frankly surprising to me because I'd have assumed it would be perfectly suited for that task.
reply