shaurya

1.2K posts

shaurya banner
shaurya

shaurya

@0oAstro

ai automation eng ee '28 @iitdelhi :: @StanfordGSB Researcher building ai agents & workflow systems

Katılım Ekim 2022
170 Takip Edilen166 Takipçiler
Sabitlenmiş Tweet
shaurya
shaurya@0oAstro·
I build AI systems for enterprise, government & hiring. Exhibits: 1. Meghdoot - gov grade grounded AI agent for official use. 2. Akira - sub 1s AI HR interview & evaluation agent. 3. Coeus - AI technical interview system w/ live code analysis. Building serious AI. DMs open.
English
0
0
4
614
Nikita Bier
Nikita Bier@nikitabier·
Every consumer-facing API client that we’ve allowed has 100x the spam reports of our iPhone app. We would need the client to send us device telemetrics, attestation, and much more. It would be Manhattan Project of API engineering. And upon achieving this incredible feat, we would be rewarded with “X Mailer: Send mass DMs and AI-generated replies”
English
86
20
860
31.4K
Wirelyss 👁️‍🗨️💫
The best thing x could do rn is allow third party clients again. It allows other intelligent cracked teams to make the X experience better and more curated, then x can learn from it and make the base app better over time (or acquire the best one lol)
Wirelyss 👁️‍🗨️💫 tweet media
Elon Musk@elonmusk

@diana_dukic What needs to be better?

English
9
6
188
43.7K
Dmitriy Kovalenko
Dmitriy Kovalenko@neogoose_btw·
fff is now built and compiled using stable rust (yes we used nightly all this time) all the platform specific optimization were rewritten either in inline assembly or std::arch
Dmitriy Kovalenko tweet media
English
1
0
55
3.4K
shaurya
shaurya@0oAstro·
@theo You prolly might not. They clearly made agents sdk for experimenting around with Claude. If they do then 😭😭😭😭
English
0
0
0
235
shaurya
shaurya@0oAstro·
@theo If I'm write you use Claude agents sdk right? It adds a lot of delay on top tho.
English
0
0
0
174
Theo - t3.gg
Theo - t3.gg@theo·
T3 Code now supports Claude. If you have the Claude Code CLI installed and signed in, you can use it with T3 Code. Hopefully the lawyers won't make us remove this 🙃
Theo - t3.gg tweet media
English
114
23
1.4K
117.4K
Melvyn • Builder
Melvyn • Builder@melvynx·
let's be honest: Yes, GPT-5.4 is smarter. Yes, it actually is; it's okay. But it's a boring model. It is limiting, lazy, and doesn't want to output my env when I explicitly ask for it. What is the purpose of this?
Melvyn • Builder tweet media
English
24
1
87
9.2K
shaurya
shaurya@0oAstro·
how did u configure it? try using one of the wrappers or using docker compose as follow ``` services: cli-proxy-api: image: eceasy/cli-proxy-api:latest ports: - "8317:8317" - "8085:8085" - "1455:1455" - "54545:54545" - "51121:51121" - "11451:11451" volumes: - /opt/cliproxy/config.yaml:/CLIProxyAPI/config.yaml - /opt/cliproxy/auths:/root/.cli-proxy-api - /opt/cliproxy/logs:/CLIProxyAPI/logs restart: unless-stopped ```
English
1
0
0
259
hwp
hwp@hwpea_·
@0oAstro @tharshan_09 @thdxr It’s pretty annoying to configure tho, for example I set it up for codex and for some reason it won’t auto name the sessions and can’t handle image inputs. Probably missed something during configuration but couldn’t be bothered given they ignore all the GH issues
English
1
0
1
260
dax
dax@thdxr·
opencode 1.3.0 will no longer autoload the claude max plugin we did our best to convince anthropic to support developer choice but they sent lawyers it's your right to access services however you wish but it is also their right to block whoever they want we can't maintain an official plugin so it's been removed from github and marked deprecated on npm appreciate our partners at openai, github and gitlab who are going the other direction and supporting developer freedom
English
209
353
6.8K
581.2K
shaurya
shaurya@0oAstro·
@stolinski i kinda dont get hype around opencode tbf
English
0
0
0
44
Scott Tolinski - Syntax.fm
I love OpenCode but "Don't rely on flat, single-color backgrounds; use gradients, shapes, or subtle patterns to build atmosphere." should not be a choice it makes for me.
Scott Tolinski - Syntax.fm tweet media
English
17
0
229
31.5K
Jonathan Wilke
Jonathan Wilke@jonathan_wilke·
Holy crap... with the @nextjs 16.2 update the next server is now using 10GB+ of memory and my MacBooks fan is running constantly (before it never even turned on). What's going on here...
Jonathan Wilke tweet media
English
60
9
445
61.5K
shaurya
shaurya@0oAstro·
@ahmed_esmail98 @thdxr I got banned from antigravity so won't recc that. Claude and coded works like a charm tho
English
0
0
0
665
shaurya
shaurya@0oAstro·
@thdxr 1password cli/bitwarden cli + claude code. they dont reveal passwords by default and u can block that too.
English
0
0
0
136
dax
dax@thdxr·
ai password manager is that anything?
English
133
1
325
46K
OpenAI Newsroom
OpenAI Newsroom@OpenAINewsroom·
We've reached an agreement to acquire Astral. After we close, OpenAI plans for @astral_sh to join our Codex team, with a continued focus on building great tools and advancing the shared mission of making developers more productive. openai.com/index/openai-t…
English
455
787
6.9K
3.6M
Hriday Kadam
Hriday Kadam@hridaykadam·
Creating a community for all selected people in YC startup school India!! Dm / comment with "YC Sus" / screenshot of selection!
Hriday Kadam tweet media
English
30
2
51
6.6K
shaurya
shaurya@0oAstro·
@paulg dont u think distribution has much much greater role in today's world given the fact that execution has never been this fast.
English
0
0
1
10
Paul Graham
Paul Graham@paulg·
Someone asked if it's a good idea to start a startup when you have nothing notable on your resume. Absolutely. All that matters in a startup is whether users like the product, and users don't care (either way) what's on your resume.
English
266
307
4.6K
167.3K
shaurya
shaurya@0oAstro·
@paulg as are tools popular before 2020. just coz llms were literally trained on them and are much more proficient. tbf i feel like this doesnt allow adoption of new technology or even new updates to existing frameworks like wide spread hallucination llms had when tailwind v4 launched.
English
0
0
1
2.2K
Paul Graham
Paul Graham@paulg·
"Anything made before 2028 is going to be valuable." — an OpenAI employee implicitly discloses their timetable
English
270
304
7.2K
968.7K
shaurya
shaurya@0oAstro·
@levelsio @stukennedydev might just be coz of the fact it is so poorly written. a better implementation like nanobot on a locked down server shouldn't really be a problem.
English
0
0
0
83
@levelsio
@levelsio@levelsio·
Another great argument for running Claude Code on your VPS server and not your laptop is its battery use "Terminal" app here is all Claude Code sessions, ignore the Claude app here I have a MacBook Pro 13" M4 and with Claude Code running even on idle my battery dies from 100% to 0% in about 3 hours, it's insane Claude Code on server via Termius SSH sucks 20x less power for your laptop
@levelsio tweet media
English
183
73
2.2K
228.3K
shaurya
shaurya@0oAstro·
look into droid btw one of THE best harnesses comparable to codex and you can feel it too while using. much better than claude for coding. my current use cases are: - codex for bug fixes or precise code edits ik - droid for planning part of things - claude for sys-admin/supervision/co-work aliter atp. factory.ai/product/cli
English
0
0
0
4