utopman

8 ms·

1.
by utopman 6d ago
I hope this is not too provocative, but where I work the legacy code is such a spaghetti mess and so actively inefficient (queries in loops of loops and plenty of things like this) that just prompt nowaday llm to "write extensive and m
2.
by utopman 3mo ago
you are absolutely right. my title is very bad. I'll update it to a very less absolute statement. sorry for that. I am now trying to sweet spot things with 27b model + tubo8 so I guess should have plenty quality context left. the erro
3.
Show HN: Best setup local LLM found for a 5090 (llama.cpp fork + turboquant)(local-llm.utop.workers.dev)
2 points by utopman 3mo ago|2 comments
4.
by utopman 8mo ago
Maybe it is just me, but : these emails are spam. Marking them as spam should be easy in a common email box nowdays. Marking these undesired emails as spam lowers email sender reputation, then finally gives real insight to the spamer soon o
5.
by utopman 9mo ago
This. I am mid career now. High level langages like js or python have a lot of bad design / suboptimal code... as well as some java code in many places. Some bad java code (it just needs to be a sql select in a loop) can easily perform
6.
by utopman 10mo ago
Then the thing becomes : Understand how to do the thing Do the thing Write down fresh documentation about the thing
7.
by utopman 10mo ago
Not sure it is your case but I think I sometimes had to solve many of them when I am in my daily task rush. My hypothesis is that I solve them too fast for "average human resolving duration" recaptcha seems to expect (I think solv
8.
Ask HN: Consumer grade hardware AI sound / music generation models
1 points by utopman 1y ago|0 comments
9.
by utopman 1y ago
This. I did not find other comments like this in this thread. 6% M2 cpu (even single core) is a huge computing budget for such a small feature. At this point I don't really understand why OP seems happy with this result (sorry). I thin
10.
Ask HN: What Happened to Noiselith
1 points by utopman 2y ago|0 comments
11.
by utopman 3y ago
This is my thought of the day about this topic I wanted to share.
12.
by utopman 3y ago
It does not seems to appear in comments yet and maybe it is a bit out of context but android devices nowdays streams almost perfectly services like geforce (video hardware acceleration is a basic feature on phones) now and it is relativiely
13.
by utopman 3y ago
I think i will do it yes. It is a small task easily doable :)
14.
Show HN: Chat GPT TUI(github.com)
3 points by utopman 3y ago|2 comments
15.
by utopman 4y ago
I have a quite kiss solution for this I gess It is a low level / harware solution that should solve th issue : a plug timer ( https://www.amazon.com/s?k=plug+timer&crid=3ORDWNDPTFJC4&spr... )
16.
by utopman 5y ago
I searched last result page on a google search with 1M results announced. On page 20 or so it is shown "display more results" then I clicked it and it still shows 1M results or so. Then I go to last page and it is only on page 40
17.
I Finnished Google Twice(youtube.com)
2 points by utopman 5y ago|3 comments
18.
by utopman 6y ago
Thank you for this constructive answer. I understand here that indeed there might have answers in NLP for my question. I also mainly understand that there are no simple answers. It's a very hard question that should be splited in many
19.
Ask HN: I wonder if message length have impact on communication efficiency
2 points by utopman 6y ago|4 comments
20.
Ask HN: Dynamic lowering “fast” graphic settings in games : Save a panda !
2 points by utopman 8y ago|1 comments