A friend of mine is looking for an entry level job in Austin. They were applying to jobs they found on job boards or to local stores/brands that they were aware of. It occurred to me that the majority of available jobs probably fall in neither category, and that the jobs they were applying to, given how they found them, were probably more competitive than average. So I asked ChatGPT to look for companies operating in Austin with either a physical office or storefront that should have entry level jobs.
The results were pretty great. It created a tabbed spreadsheet of over 250 such companies, labelling each by company size, region, and job categories. It even assigned them priorities - which should be applied to first. I'd guess 250 is only about 10% of companies that met the criteria that I gave, but with only one shot, it was already way better than pouring over a job board imo.
What's something you used an LLM for recently that wasn't work related?
The last thing I did was upload all the condo bills from the building where I've lived for the past year, since the new trustee took over, to Notebook LLM. The cost of the condo had increased significantly, and I asked him if there were any overcharges or any signs of corruption. I was shocked by everything he found. I posted the results as questions to the condominium's WhatsApp group, questioning the trustee, and it's caused quite a stir—so much so that a meeting has been scheduled for August 5th!
I'd be curious if this actually resulted in a successful job find, coz that would be pretty great. But were they mostly restaurants/retail stores?
In addition to my work, I often teach at my church. I use AI to help me prep bible study. I'm very cautious not to let it interpret any passages for me, but it's useful if I can't remember the exact words of a verse and I forgot exactly where it is, but I remember the general gist of it... it can help me pinpoint which verse of the Bible it was. It's also good for summarizing the various scholarly opinions on particular passages.
We don't know yet! This was literally yesterday.
A good portion of them were, but they were consistent with the criteria. I also re-prompted later to look for tech startups with 50-200 employees (about the right size for fun, chill, and stable).
I often use them to search in place of google. I was wondering something about F1 the other day and I went into a rabbit hole about F1 safety standards even down to all the gear drivers wear and in what order they put it on. Haha
This is my most common usage too.
I use it a ton for translations.
Cooking aid + house & car maintenance questions so that if I don't DIY, I at least know enough that the professional I get a quote from won't make shit up
Cleaning up Jellyfin library with OpenCode.
Filenames, additional files, metadata, title metadata, actors and director and genre etc.
I rarely use LLMs for work. I find myself using it to compare products on the shelf at the store.
inspecting the coinkite firmware
I also know somebody who's using it to navigate some family drama
I wanted to see what Obscene Selene looked like if she were bald. Grok completely misunderstood my request and gave me an ancient Greek goddess and personification of the Moon, completely naked. It wasn't what I wanted but I'll take it.
I plug in my exercise notes into various LLMs for analysis and insights. It's helped build structure to my routine and led to some useful refinements.
I've been pointing Sol at a folder containing Liber Primus to see if it can solve any of the 56 unsolved pages from Cicada 3301's final(?) puzzle.
I had Fable attempt the same and it fucked something up so bad that I cancelled immediately.
https://en.wikipedia.org/wiki/Cicada_3301
I had it research agri stuff for a friend Tuesday night. I'm pretty much pleased with the result.
I saw that one!
Yeah! I feel that the current generation LLM (basically the second generation of models trained to execute tools rather than trying to be a smarty pants on their own) is getting pretty good at
search, compare, incorporatepatterns.The other day my codex'd GLM was without me asking explicitly mapping known vulns to code when reviewing a massive release milestone on a library. I merely instructed "all details must be referenced" and "leave no stone unturned" among acceptance criteria, which apparently can lead to employee-of-the-month behavior. So I guess GLM is going to be just that, lol.
AI is used in all areas of life, but I'd never heard of it being used for job searching. Good for your friend, and thanks for sharing with the community!
Sites like IMDB are terrible so it's great for finding old movies for the queue
Garden planning and plant/bug identification
Basically search displacement, but it really shines when those searches involve pairing or contrast to other searches.