Atilay

777 posts

Atilay banner
Atilay

Atilay

@AtilayCandan

AI | Coding | Gym | Chess 👉 Get a comprehensive analysis of your website in 5 mins 👉 https://t.co/AvhyutnS26

United States Katılım Mart 2024
126 Takip Edilen87 Takipçiler
Sabitlenmiş Tweet
Atilay
Atilay@AtilayCandan·
🚀Recently launched a Website Analysis Tool for businesses that generates 20-page audit reports with detailed scores and actionable recommendations on 39 different metrics. Over 120 businesses already generated their reports on the first day. sitalyze.com
English
3
0
8
1.7K
Atilay
Atilay@AtilayCandan·
Rest In Peace Daniel Naroditsky My favorite chess personality of all time. He was the kindest and the best teacher. I spent countless hours in the last 4-5 years watching his videos and streams. I also played him in one of his speedrun videos and he crushed me in 27 moves:
English
2
0
4
3K
Atilay
Atilay@AtilayCandan·
@GabrielePacioc1 @mrexits Back then you could leverage a small amount of money to get a lot of stuff done and now it's essentially free lol Ideas matter more than ever
English
0
0
1
11
Atilay
Atilay@AtilayCandan·
@gnukeith I feel like Youtube is really pushing this ad nonsense too much There will come a point where people will say that's enough and move on to a different platform altogether
English
2
0
1
300
Keith
Keith@gnukeith·
YouTube: Everyone is using an Ad-Blocker how do we fix this problem??? Employee 1: How about we play ads next to the video that the user can skip in 5 seconds so that if they are interested they can check it out and it isn't intrusive? Employee 2: Unskippable ads YouTube: Awesome!! Employee 2 you get a raise! Brave: The Lion does not concern itself with YouTube ads
Dexerto@Dexerto

YouTube is testing out longer unskippable ads that are set to appear more often for viewers

English
19
19
217
12.5K
Atilay
Atilay@AtilayCandan·
Sam Altman's blog post today highlights how dangerous social media algorithms are, calling them 'misaligned AI'
Atilay tweet media
English
1
1
1
293
Atilay
Atilay@AtilayCandan·
@sama show us some benchmarks against claude 4 and gemini 2.5 pro
English
0
0
8
532
Sam Altman
Sam Altman@sama·
o3-pro is rolling out now for all chatgpt pro users and in the api. it is really smart! i didnt believe the win rates relative to o3 the first time i saw them.
English
446
405
6.8K
914.2K
Atilay
Atilay@AtilayCandan·
@levelsio I think this would work really well if you could only create one personal account with your biometrics. Idk how you could implement this safely though, and how you could get people to trust you enough with their biometrics. Definitely a future billion dollar idea tho
English
0
0
0
23
@levelsio
@levelsio@levelsio·
Yes so we need biometrics when posting content to verify it's human But even then people will just use their biometrics to AI post But less
Esa@esa_was_taken

@levelsio @X It's interesting that the arms race between attacker and defender on this front is already in full swing and it's only going to get harder and harder to figure out whether someone is human the internet is kinda fucked for a while

English
65
5
137
83.4K
Atilay
Atilay@AtilayCandan·
@IterIntellectus Even intern devs know how important accessibility is lol You can't even read anything. How are they going to release this to hundreds of millions of people?
English
0
0
30
2K
vittorio
vittorio@IterIntellectus·
steve jobs is in heaven crying
vittorio tweet media
English
148
659
13K
733K
dax
dax@thdxr·
trying to have an open mind but i can't see this as not ugly
dax tweet media
English
68
7
580
42.4K
Atilay
Atilay@AtilayCandan·
@codyschneider Doesn't this also mean competition is tougher than ever?
English
0
0
1
207
Cody Schneider
Cody Schneider@codyschneider·
cost of marketing going to zero cost of code going to zero basically free reach with social basically free to build an app
English
52
16
539
94.5K
Atilay
Atilay@AtilayCandan·
@levelsio @yacineMTB @elonmusk The pinned post notifications from communities is also incredibly annoying I see them a couple times a day at least and it's always just advertisement
English
0
0
0
95
@levelsio
@levelsio@levelsio·
People gonna hate me for this but this would solve a large majority of AI replies on X AI replies usually don't pay for X @yacineMTB @elonmusk
@levelsio tweet media
@levelsio@levelsio

This has to be @X's #1 priority I can't keep blocking every AI reply anymore, it's become too much I did used to hide non-Premium users, which might help, but I can't find the setting anymore?

English
131
11
625
126.4K
Atilay
Atilay@AtilayCandan·
Other big companies: We have to be the first to build AGI. AI is the future! Apple: Liquid glass from 2010 go brrr
English
0
0
2
75
Atilay
Atilay@AtilayCandan·
@zack_overflow Yeah I guess context size/quality will remain to be an issue in the foreseeable future. Also, how do you get to 18,000 lines lol? I get anxiety after I reach like 300 lol. That file would crash Cursor's $10B servers.
English
1
0
3
779
zack
zack@zack_overflow·
I just split up an 18,000 line file in Bun Not because it makes it easier for humans to read But primarily because it makes it easier for LLMs to read
zack tweet media
English
18
5
365
38.3K
Atilay
Atilay@AtilayCandan·
@theo It honestly feels like AI models are just so much ahead of the infrastructure we currently have. You'd think that a problem like this would be solved by now with everyone building 69 AI apps a day, but everything is slop I guess.
English
1
0
6
737
Theo - t3.gg
Theo - t3.gg@theo·
Okay so in-browser AI is still useless, got it.
Theo - t3.gg tweet media
English
81
13
1.2K
92.6K
Atilay
Atilay@AtilayCandan·
Is it just me or does it feel like we only get banger AI models like only twice a year? Last year: -o1 -claude 3.5 sonnet This year: -gemini 2.5 pro -??????? Sure each iteration improves on the last but some models are just way more forgettable even if they are better.
English
0
0
1
92

Keşfet

@GabrielePacioc1 @mrexits @gnukeith @sama @levelsio @IterIntellectus @thdxr @codyschneider