Extremely misleading railway isn’t a vibe coding platform what an awful title.
completely understand that they had a fuck up not a big railway lover but what an awful title
The vibe-coding platform said all of its customers should revoke and rotate their private keys, after hackers were able to decrypt some customer secrets.
spr.ly/6017BHkxEb
i genuinely just might retire as a software engineer.
i really need to learn to play my gibson so i can be a guitarist smh
who touched prod today?
“Who touched prod?” Now with filters
Changelog #0312
• Activity feed filters
• Guided IaC migration
• Cron jobs in the CLI
• Fixes and improvements
railway.com/changelog/2026-1…
182 issues closed and 40 PRs open, all thoroughly tested and ready to merge tomorrow. All in one chat: Opus 5.5 instructing GPT 6.1 Sol and Haiku 5.5 underneath it. T3 Code is such an amazing product.
@theo you cooked
half of it in the mobile app to.
Max retweeted
🫡
Introducing tsrs: a Rust port of the TypeScript compiler 🦀
We are not forking TypeScript!
◆ In close collaborating with the TypeScript team
◆ Up to 8x faster than TS 7.0 & ~70% less memory
◆ Runs in WASM, so it works in your browser & on Workers
◆ Soon part of the Oxc toolchain
npx tsrs
blog.cloudflare.com/typescri…
Max retweeted
Grateful to @deepseek_ai for releasing DeepInfer, initially built primarily on @vllm_project, under Apache 2.0 and crediting the open-source community!
We've worked hard on DeepSeek V4.1 Flash inference efficiency and will keep improving with ideas from DeepInfer! Stay tuned 🙌
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
by producing the amount of units as the name suggests
New US partners like Perryman and Qnity Electronics have helped to make iPhone Duo and iPhone 18 Pro possible. apple.com/newsroom/2026/10/a…
Big props to @theo and his team. holy fuck i just got into t3 code and it is extremely good. nothing is competing for my attention and it has this strong philosophy of it just fucking works. it's fast it's light i downloaded the mobile app scanned a code and now i can start chats from my phone.
it is genuinely incredible the amount of attention to detail the sub agent view is perfect the pull request flow is orgasmic and i also love how i can see costs and usage limits.
big fucking props i originally thought this would be slop but it feels like and i can't believe i am saying this. a better version of the codex desktop
i’m doing a lot and can’t even reach 30% of my weekly limit lol
Max retweeted
I'd much rather work with someone who flexes much they accomplished with the fewest amount of tokens vs how many tokens they spent, and I sense this is a major trend in general
How we quantify this as an industry is still TBD, probably a major new set of benchmarks
crazy to think you have to request access to a model that’s supposed to protect you.
We need more open source models specifically for defenders.
Check out chatgpt.com/cyber to get Daybreak for individuals and openai.com/form/enterprise-t… for organizations.
We're here to help.