[dead]
Building LLM Applications for Production
11–20 of 99 posts
Re: Building LLM Applications for Production
#12I like a lot of the LLM use cases mentioned here. A couple more are:
- conducting literature reviews (stay sane while researching LLMs!)
- Talking to textbooks / AI teaching assistants
- language learning with a companion tailored to your level and interested
LLMs are so hyped and written about these days that it would be hilarious if the next version of GPT trained on todays internet would be biased towards praising itself
Re: Building LLM Applications for Production
#13So now almost all the low hanging fruit programming books have instantly become redundant and off-shored to ChatGPT, and will stay on the shelves to collect dust. New here comes the race to create prompt engineering books and courses in. 24 hours to sell to other AI bros who think that they are prompting it wrong, not prompting hard enough or the prompting the wrong way.
> here comes the race to create prompt engineering books and courses in. 24 hours to sell to other AI bros who think that they are prompting it wrong, not prompting hard enough or the prompting the wrong way.
That’s already been happening for a couple of months now.
Hilariously some of the AI bros that sell the AI prompting video lessons do not put effort into quality of the material of the videos. Instead they make use of the AI themselves to shovel out low quality garbage, which they then package as expert advice and sell to others.
Re: Building LLM Applications for Production
#14So now almost all the low hanging fruit programming books have instantly become redundant and off-shored to ChatGPT, and will stay on the shelves to collect dust. New here comes the race to create prompt engineering books and courses in. 24 hours to sell to other AI bros who think that they are prompting it wrong, not prompting hard enough or the prompting the wrong way.
> here comes the race to create prompt engineering books and courses in. 24 hours to sell to other AI bros who think that they are prompting it wrong, not prompting hard enough or the prompting the wrong way. That’s already been happening for a couple of months now. Hilariously some of the AI bros that sell the AI prompting video lessons do not put effort into quality of the material of the videos. Instead they make…
Welcome to the world of tomorrow!
Re: Building LLM Applications for Production
#15It's inaccurate to attribute all of these use cases to "LLMs" in general when currently only 3 or 4 of the best models can do all of them well. Especially the ones that involve writing code or highly technical instructions. It's OpenAI plus maybe one other model from another group, but just barely.
Re: Building LLM Applications for Production
#16Earlier quoted context omitted.
OK, now extract this sentiment to the whole of academia. By the time the average syllabus starts being taught at an academic institution, it can be several years out of date, and by the time you finish it, it's already five years out of date. Takeaway: there's a lot wrong with the existing educational system and how we pass on actionable theory.
I actually no idea how one could teach stuff like Bag of Words naive bayes classifier etc for a whole semester and charge $4000 like most universities- with a straight face
You do still need a grounding in calculus and the fundamentals, even where dated, to know what's going on behind the APIs and models.
If you're OK using an ORM with no relational database or SQL knowledge (as a parallel) then sure it makes no sense.
Re: Building LLM Applications for Production
#17Are there any great examples of cost efficient LLM application deployment at scale?
Maybe look at what AI Dungeon has done.
Re: Building LLM Applications for Production
#18I like a lot of the LLM use cases mentioned here. A couple more are: - conducting literature reviews (stay sane while researching LLMs!) - Talking to textbooks / AI teaching assistants - language learning with a companion tailored to your level and interested LLMs are so hyped and written about these days that it would be hilarious if the next version of GPT trained on todays internet would be biased towards praising…
Here's hoping that my message about the discourse about the praise of GPT leading to a self-praising GPT itself leads to a more introspective GPT, is used as training data and results in a more introspective GPT, which I suppose is me praising GPT, which will result in a more self-praising GPT...
Re: Building LLM Applications for Production
#19The novelty is wearing off and the reality of parsing hundreds of TBs into hundreds of GB memory blobs you can query by the Kb is setting in.
Re: Building LLM Applications for Production
#20It's inaccurate to attribute all of these use cases to "LLMs" in general when currently only 3 or 4 of the best models can do all of them well. Especially the ones that involve writing code or highly technical instructions. It's OpenAI plus maybe one other model from another group, but just barely.
Are there any models aside from OpenAI’s that can handle large prompts with task breakdowns? I haven’t tried the Anthropic stuff, but every flavor of LLama and other open source models do not seem capable of this.