Sanchit

1.1K posts

Sanchit banner
Sanchit

Sanchit

@sanchitcodes

i love designing and building scalable products | follow to watch me become something from the ground up :) https://t.co/t7ioiYMM3d

building tech Katılım Aralık 2024
53 Takip Edilen65 Takipçiler
Sabitlenmiş Tweet
Sanchit
Sanchit@sanchitcodes·
may this pin serve as a reminder for me to continue grinding i will unpin this when i make it Status at the time of writing this: $0 MRR, 41 followers Goal: $10K MRR, 100k followers follow me on my journey, thank you and peace, friend✌️ 🧵 thread (will update monthly)
English
1
0
3
584
Sanchit
Sanchit@sanchitcodes·
@OfficialLoganK wow is this a tease for the upcoming gemini model having a really low hallucination rate? that was one of my top 3 biggest problems when i use gemini models especially flash inside of kilo. loved its speed but it just refused to do work and hallucinated a lot
English
0
0
2
894
Logan Kilpatrick
Logan Kilpatrick@OfficialLoganK·
Why don’t LLM’s just tell you when you are asking a question / doing something that is out of distribution?
English
262
48
1.6K
154.1K
Sanchit
Sanchit@sanchitcodes·
@YashHustle_22 switched froml cc to getting a gpt sub and using it inside of kilo mainly cuz 5.5 is just leagues better rn at coding. also the rate limits were really horrible with claude and i love the fact that i can use my gpt sub anywhere i want
English
0
0
0
10
Yash
Yash@YashHustle_22·
Vibe coders, What made you switch from Claude to Codex?
English
22
1
16
1.1K
Sanchit
Sanchit@sanchitcodes·
@cursor_ai man i wish u all give api access as well for third party tools. i would love to give it a shot and probably make it my daily driver inside of kiloclawq butttt..
English
1
0
0
3.7K
Cursor
Cursor@cursor_ai·
Introducing Composer 2.5, our most powerful model yet. It's more intelligent, better at sustained work on long-running tasks, and more reliable at following complex instructions. For the next week, we’re doubling the included usage of the model.
Cursor tweet media
English
669
956
9.7K
12.1M
Sanchit
Sanchit@sanchitcodes·
@synthwavedd only flash or both flash and pro?? excited for pro. i hope it is very comparable to gpt 5.5 and is good at doing agentic work since 3.1 was really bad in kilo with it and hallucinated a lot in antigravity as well
English
0
0
0
467
leo 🐾
leo 🐾@synthwavedd·
Gemini
Indonesia
16
3
280
10.2K
Sanchit
Sanchit@sanchitcodes·
@matanSF hello im changing my name to AIden Factoria
English
0
0
1
405
Sanchit
Sanchit@sanchitcodes·
@mark_k @GoogleDeepMind so excited can't wait for it..i just hope that they keep the pricing the same or slightly lower it down because i'd love to continue using them as my daily driver inside of my kiloclaw setup
English
1
0
1
1.2K
Mark Kretschmann
Mark Kretschmann@mark_k·
This week is Google I/O. Many interesting AI releases from @GoogleDeepMind are expected: - Gemini 3.5 Flash - Gemini 3.5 Pro - Veo 4 - Google Omni (multimodal video model) And much more... 💫
English
58
47
1.1K
52.9K
Sanchit
Sanchit@sanchitcodes·
@OfficialLoganK love seeing what people have been talking about 3.2/5 flash inside since some of them are being routed to it. i've been hearing it's not lazy and that was one of my biggest pain points with gemini models when i tried them inside of kilo..can't wait for a pro preview either
English
0
0
1
1.7K
Kath Korevec
Kath Korevec@simpsoka·
What other Codex papercuts should we fix?
English
367
5
287
34.9K
Sanchit
Sanchit@sanchitcodes·
@VisualFidelity yea fh6 makes me want to have a console but it is what it is ig..cant wait only 2 more days to go!!
English
0
0
2
110
Visual Fidelity
Visual Fidelity@VisualFidelity·
Forza Horizon 6 makes me wish I had a Series X for the first time in 2 years since I sold it. While I'll play on my PC in a few days, my PC is in my office and I prefer to play on the living room on my 77 OLED at night before bed. Series X at $649 is way overpriced. If I was $399 for a discless version I'd pick one up again. Once it releases on PS5 Pro I'll pick it up there Exclusives are good for Xbox and lower hardware prices might win back some market share.
GIF
English
25
1
77
10.4K
Theo - t3.gg
Theo - t3.gg@theo·
@LinusEkenstam Where did I like my own post? I liked things from orders that I also retweeted?
English
3
0
53
2.1K
Linus ✦ Ekenstam
Linus ✦ Ekenstam@LinusEkenstam·
Stop deviating from the real problem. Posting something that goes viral, to then be punished by 93% reduction in reach within 48h or candidate pool refresh. Here, me hunting for undisclosed paid posts. Enjoy
Nikita Bier@nikitabier

@LinusEkenstam You do many undisclosed promotions. No one trusts you.

English
38
2
83
134.7K
Boris Cherny
Boris Cherny@bcherny·
@sickdotdev 👋 was this using Claude Code? If you wouldn’t mind running /usage and pasting the full output here, I’d be happy to help debug. We’re also actively working on making it easier to self-serve to see what exactly is using up your limits.
English
135
7
1.4K
253.6K
Sick
Sick@sickdotdev·
Opus 4.7 sure was fun for about 15 minutes
Sick tweet media
English
156
22
2.4K
404.6K
Thomas Ricouard
Thomas Ricouard@Dimillian·
It’s time. Now you’ve tried our Codex mobile, tell me the top missing feature for your workflow! And yes, we’re aware of the current bugs and shortcomings. We’re working hard on it!
English
290
13
542
38K
Sanchit
Sanchit@sanchitcodes·
@ChatGPTapp still not for windows btw. 2026 btw. no linux app either btw.
English
0
0
0
58
ChatGPT
ChatGPT@ChatGPTapp·
Touch grass and leave your laptop—Codex is now on your phone. Now in preview on the ChatGPT mobile app.
English
186
173
2.6K
161.9K
omni
omni@omni1896837·
@sanchitcodes @sama I think windows don’t have the necessary APIs for computer use (e.g apps can’t be controlled via code) Not sure though. Hoping they will give it some care soon.
English
1
0
0
42
Sanchit
Sanchit@sanchitcodes·
@omni1896837 @HarshithLucky3 i dont have a lot of codex usage to spare, if u have the limits could u ask it to make a .md file about what all it did step by step? thanks a lot!! are u able to invoke skills and mention files and stuff thru ur phone?
English
0
0
1
53
omni
omni@omni1896837·
Working now! I mostly got codex to do it, took about an hour so the main things are - - Enable feature flag in the codex app - When the iOS app is sending data to the codex app on windows, it assumes it’s communicating with a MacOS pc, so things like the path names need to be correctly handled when the request is coming in. There might be some other issues. I used GPT 5.5 xhigh with /goal to get it done
English
3
1
0
175
omni
omni@omni1896837·
@sanchitcodes @HarshithLucky3 Kind of. I can message already existing chats, but unable to create new chats from the app. Debugging at the moment.
English
1
0
0
101
Sanchit
Sanchit@sanchitcodes·
@OpenAI "soon" just like computer use which was coming soon as well??? or atlas?? it was my mistake ig waiting for the codex app but nvm. so stupid, and still no linux app 🤡
English
0
0
3
983
OpenAI
OpenAI@OpenAI·
Rolling out today as a preview on iOS and Android in all supported regions. Support for connecting your phone to the Codex app on Windows is coming soon. openai.com/index/work-wit…
English
159
159
1.7K
313.8K
OpenAI
OpenAI@OpenAI·
You've been asking for this one... Now in preview: Codex in the ChatGPT mobile app. Start new work, review outputs, steer execution, and approve next steps, all from the ChatGPT mobile app. Codex will keep running on your laptop, Mac mini, or devbox.
English
1.6K
2.6K
21.8K
4.5M