Ask Lemmy

I've been using Google's Gemini to write cover letters for job applications. Just plug in the job description, do a little proofreading and tweaking, and boom. It's made the process so much easier for "personalized" cover letters.

I use Kagi, they provide access to all the main models in a chat interface and have a mode that feeds search engine results to them. It's mostly replaced search engines for me. For programming work I find them very useful for using unfamiliar tools and libraries, I can ask it what I want to so and it'll generally tell me how correctly. Importantly, the search engine mode has citations. $25 a month, but worth it.

I am not paying for pro version on monthly basic. I am using openrouter.ai, I load $5 and pick the bot when I need to use.

Still have $3 after few month.

AI pay as you go cost doesn't cost that much

Once upon a time I had it writing summaries of some of the comic series I was selling on Whatnot. Did that twice. Haven’t used it for anything else.

I use Gemini as a replacmet for google search. Still kinda shit but everything else wants money or bans VPN users.

I run Mistral locally for anything personal or fun but I need a new GPU. My 1080ti is finally showing its age.

Not a daily routine.

I've been doing some visual character stuff for my fictional story!
It's nice to actually see some characters visually for the story - it adds to the motivation to work on it 😁

I don't unless I am trying to make my code more efficient, or want know how to do simple programming task or get some additional info on something I don't really understand.

I'll typically only use it for language and coding problems.

Synonyms, word for xyz, how can I make this sentence more clear.

But if I can't find anything on Google I'll ask it other questions.

I use ChatGPT 4 and 3.5 via the api sometimes.

I've recently started using Claude Opus which i prefer for coding and maybe in general. Apart from the crappy message limit and lack of stop and edit buttons. I think it might stick with it and cancel gpt.

In general i use Ai to help with my job of general it guy, marker, admin, data entry etc. i use it to proof read, to get ideas, to transform data, to code it or to comment code, translate, transcribe, to bug fix. I always feel i could be making better use of it in my day but it is transformative for the stuff it does.

For the API I'm using a workflow for Alfred on Mac that allows me really quick key presses access to prompts and questions on selected text. I love it.

Ollama running the dolphin-mistrial. Its been neat using it with the continue plugin in vscode, nothing life changing though. I think with further embeddings and RAG over curated data sources would go a long way to make it more useful for me.

I work as a research economist and use half a model zoo and APIs regularly and have even written a small R package to work with LLM Apis. ChatGPT Pro in the interface for programming questions (helping me to write or document my R code, for example I have used it to easily translate tax laws into R functions to make microsimulations [of course i double check them]). Anthropic's or Groq's API's to process large amounts of documents fast (for examples creating JSON-lists about papers to make them more easily searchable). I have one small script that has many useful prompts that really helps me to rephrase texts (e.g. "Please rephrase this paragraph to be more clear and concise. Give me {n} Versions. {paragrph}") , which I have included into my browser. I used Stable Diffusion to generate images for my Christmas Cards.

usually don't if not coding in Roblox

I use the Bingilator to create images for invites to my weekly donut meetings.

I use it to write Ansible scripts; simply because Ansible sucks so fucking hard and I hate to do it myself.

For programming I use it a lot. Though, it's because the book hasn't been helpful and the professor is non-existent

Nothing, I’m not a loser