m3kw9

6 ms·

1.
by m3kw9 4d ago
I don't think with the models we have, you shouldn't need to even write a single line of code. What you need is to prompt it right, check for over engineering, check for edge cases, test it very thoroughly. That means don't
2.
by m3kw9 4d ago
is this like Aider's repo map?
3.
by m3kw9 4d ago
what if the production code base was made by mostly by Anthropic models?
4.
by m3kw9 4d ago
They are going for IPO, don't be naive
5.
by m3kw9 6d ago
whats the practical use of this?
6.
by m3kw9 6d ago
I'm using Codex, Gemini etc, they all have desktop apps and have a plan, how do i use SWE-2? Thats is a problem they have. I'm not about to switch out my workflow and plans with a shiny LLM that looks benchmaxxed and graph maxxe
7.
by m3kw9 6d ago
Good, C# is one of the worst languages there is, hopefully they turn it in to tier 18
8.
by m3kw9 6d ago
They are admitting React was a shtty choice for a mobile app if they have a choice.
9.
by m3kw9 7d ago
App developers gonna have extra work
10.
by m3kw9 7d ago
prob tech and UX issue. FaceID can't do side ways, and people opens side ways half the time. Maybe there are also functions unlockable wihtout clamshell opening. can't put 5 faceID on each side and facing every which way.
11.
by m3kw9 7d ago
^ knocking it before anyone on earth tried it- minus apple
12.
by m3kw9 7d ago
why not? lets not talk about it, people like this gonna say we are highlighting a new feature on a non-pro model
13.
by m3kw9 7d ago
They expect vibe coders to use their models only lol
14.
by m3kw9 8d ago
give them 1 dollar each run, and maybe you will make 2 bucks on the 100,000th run.
15.
by m3kw9 9d ago
so they allow auto drive in non-approved areas?
16.
by m3kw9 10d ago
they made the pic look like it can be installed on an intel motherboard.
17.
by m3kw9 10d ago
1000 years means it's only good for archival purposes, or are people gonna bank on when they pass it on to their loved ones, they'd keep more and more junk instead of using a new medium?
18.
by m3kw9 10d ago
it's all about money, they write some attractive stuff, they get some views, likes follows, maybe someone will say "this dude is smart". but most times you get idiots liking the stuff, and likely detrimental because a smart
19.
by m3kw9 11d ago
An openai/claude engineer would come in and douse this crap
20.
by m3kw9 11d ago
input tokens are cheap, is most of your task reading? And can flash 2.5 compress that read and give you a reliable output?
21.
by m3kw9 11d ago
same with software, it came fast, but we was ready. The leverge is high but it still does not have much "taste" in design. Ask it to design something, they enter a pattern. They can get it right but is the same pattern, if it&#
22.
by m3kw9 11d ago
would be trivial to just create a message board somewhere completely obscure to anyone
23.
by m3kw9 11d ago
all info has a delay, it does not have to match your tempo.
24.
by m3kw9 13d ago
efficiency per intelligence is the benchmark i look at the most, as that allows the most use by most people.
25.
by m3kw9 13d ago
They have a lot of moat, i'm not sure what youa re talking about. Only amatures are using Qwen, open source stuff that is 3-8 weeks behind. Plus OpenAI has some verticals that keep people in there.
26.
by m3kw9 13d ago
arc-agi3 is meaningless to most people. I'm not gonna look at the tests and see how hard it is. The actual test we look at is terminal bench, thats where software is being accelerated and closer to where rubber meets the road
27.
by m3kw9 15d ago
probably uses it extensively to convert between LLM working on a doc and exporting it out to excel, word etc
28.
by m3kw9 15d ago
I ain't wanna see anymore websites with "The SAAS that actually [italics]Works[\italics]"
29.
by m3kw9 16d ago
I'd think AI can help him even out the score a little, or a lot.
30.
by m3kw9 18d ago
A bit of perfectionism helps with esp UI bugs. A little flash, slight unusual pause, trying different ways to break the code..

More ›