Hacker Newsnew | past | comments | ask | show | jobs | submit | another-reader's commentslogin

Any pointers on creating similar tools? I had an idea for an AI-backed analyses tool (not for stocks) but not quite sure how to go about it. Even if it's just a pointer to some page that does an ELI5 I'd really appreciate it.


Depending on how much time you have, the first thing I would recommend is to watch Karpathy's 'deep dive into LLMs' video (https://www.youtube.com/watch?v=7xTGNNLPyMI). It is 3.5 hours but it will give you so much instinctive understanding of how these things work and will likely save you a lot of time.

Purely from a technical standpoint, things aren't that difficult really. I'd recommend get yourself an API account with all major providers (because each frontier LLM evidently has their own strengths and weaknesses) and start experimenting with your data. The feeling you'll get when you first get an API response will keep you going! :)

Having said that, a caveat is, LLMs do hallucinate from time to time. If your data needs accuracy, definitely don't do the "throw everything at LLMs and let them do the work" but I'm sure you'll figure it out from trial and error. Good luck!


> I'd recommend get yourself an API account with all major providers (because each frontier LLM evidently has their own strengths and weaknesses) and start experimenting with your data.

Wouldn't it be better to use something like openrouter.ai which gives access to lots of different endpoints? In my experiments that is what I am using, wondering if there are good reasons to go to direct API usage.


Yeah Openrouter is generally better. Only exception is when some API is launched but not yet on openrouter, rare though


Will check it out. Thanks so much!

I'm pretty familiar with how LLM's work, it was more of "how do I create sites/apps" with LLM/AI backends. Will check out the youtube. Thanks again.


I opened up File Explorer I and I see I made a full dump of Evernote on 30 July 2019. I was a paying customer for a few years but decided at the time to dump it before it was too late.

To this day, the one thing I truly truly miss from Evernote was the web archive. They had the best at the time, and I have yet to see it matched.

So, it seems from the thread that I should give it a try again?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: