Maxion

5 ms·

1.
by Maxion 7d ago
It literally is how they work. They look for the next most probable token. Always keep that in mind when writing a prompt or reading output. If you don't supply it, it will fill it in with the most obvious stuff.
2.
by Maxion 8d ago
I think it's more important now than it was last year to differentiate pure yolo vibecoding from "AI assisted engineering" or "AI engineering", I.e. deliberate and careful use of AI to speed up coding but without cr
3.
by Maxion 8d ago
AskJeeves really was ahead of its time with its name and branding.
4.
by Maxion 8d ago
Oh wow! Now that's a cool site!
5.
by Maxion 12d ago
I'm pretty convinced that manual coding will be done by late 2027 for at least webdev. At that point the optimal workflow will be coders controlling agents. Not too far after it will be coders controlling agents who control agents who
6.
One Gets Through – Missile Command, but you launch the missiles(onegetsthrough.com)
49 points by Maxion 12d ago|10 comments
7.
by Maxion 12d ago
The big manufactureres pay or just simple acquire the leaks.
8.
by Maxion 13d ago
They're used pretty often, so not really. The KYC providers anyway wouldn't code anything like that.
9.
by Maxion 13d ago
Breach yes, but if they cannot 100% sure identify if your data was given out falsily, then they cannot say. They're not allowed to disclose that they provide your information to LE. So they can only inform you directly if they're
10.
by Maxion 13d ago
In the countries you are licensed in you are legally required to reply to law enforcement requests. In most places there is no official channel for this. It is literally stuff like LE@Fintech.com. Emails come from all over and random domain
11.
by Maxion 13d ago
You can't, really. Banking legislation does not require them to tell you.
12.
by Maxion 13d ago
Yep, the KYC provider keeps them.
13.
by Maxion 13d ago
I've done that as well and this is what most of those do look like.
14.
by Maxion 13d ago
.gov is a US thing, and not even all US agencies use .gov ending emails.
15.
by Maxion 13d ago
I've processed government requests at a FinTech before. Some are pretty good and there are bespoke channels for them so that you can be sure their genuine. Other are literally random emails you get that you are required to reply to, ma
16.
by Maxion 13d ago
It's one of these things that people do think is cool but it also does limit returns and resale.
17.
by Maxion 14d ago
This will take a while for most people to internalize, but now with GPT-6-Astra we are basically in the age where as a developer you control agents to write code for you. This will be further cemented by the next generation of models. I fea
18.
by Maxion 14d ago
I searched for this a few months ago and found some mentions of this bug, but yes it affects me too. Can stall up to 20 seconds+ sometimes. Chrome is fine.
19.
by Maxion 17d ago
I find it sad that they're doing it all pre-recorded.
20.
by Maxion 21d ago
The whole architecture of current LLMs just is not conductive to AGI, it's just marketing hype. In my opinion current LLMs aren't even AI, they're just fancy machine learning algorithms
21.
by Maxion 22d ago
The comments on this issue are unbearable.
22.
by Maxion 22d ago
Proportional decrease is not as useful in a climate change sense than actual decrease in co2 emissions. A grid that is highly sustainable with lots of wind and solar, but fossil peaker plants will have a lot more potential co2 savings in th
23.
by Maxion 22d ago
A lot of fossil power stations cant shut down for a few hours, they have to keep themselves on.
24.
by Maxion 22d ago
I wanted some banana plugs for my speakers, as well as I needed some longer speaker wire. I kinda wanted it for the weekend so I could watch movies with some friends. I didn't want to order online as that wouldn't arrive in time.
25.
by Maxion 22d ago
I will happily shout AGI from the rooftops the day I can turn on voice mode in ChatGPT and have the model calm down my toddler for a tantrum, or keep him from opening all the bananas instead of eating one. People in this thread arguing abou
26.
by Maxion 22d ago
Bingo, LLM architecture just does not lend itself to becoming AGI. They can get really good, sure, but they will always struggle with novel input and scenarios. The more training data that is shoved in to them, the more they'll seem
27.
by Maxion 23d ago
And Grok
28.
by Maxion 24d ago
I wouldn't say they're at threat from AI. Consultancies are most of the time there to partially offload work from the main company, and also to serve as someone to blame when something does not go well. It's rarely about the
29.
by Maxion 29d ago
Those books are definitely in the training set if the model can mention them, but why exactly these terms are so commonly used is not necessarily because the book is in the training material. The model does not itself know how its trained
30.
by Maxion 29d ago
Yep, for those of us who were doing "agile waterfall" PMs often talked about shipping versions too.

More ›