@Tech

@DrewHintz Great article

This is the way: “For credentials, we use domain-scoped secret injection at egress. The model and container only see placeholders, while raw secret values stay outside model-visible context and only get applied for approved destinations.” openai.com/index/equip-re…

@nnwakelam @hacker @elonmusk How are they attempting? Curious

X is still attempting to take "@hacker" from me. If anyone knows @elonmusk or someone at X I would appreciate if it could get a second look.

I read the MIT paper on Recursive Language Models (RLMs). Its a clever wrapper for LLMs to handle long contexts via a REPL chunking, summarizing, recursing. Seeing post on X calling it a paradigm shift, I had to read it again to see if I was missing anything.. Its essentially agentic workflows + basic text ops from CS101. Not sure how many folks on X actually read the paper. A few onions layers away you can align this to heuristics vs semantics.

My 2025 OpenRouter Wrapped is here! 🏆 Top 1% of users. #OpenRouterWrapped

@IamKyros69 Is this filtered? I built out a KYC mobile app, the Face ID mesh didnt look like this. It also looks like this was taken on the front camera, how did you get the true depth camera to work and the IR sensor to record?

Apple's Face ID unlocks your phone using 30,000 invisible dots. This is what it sees...

Whiteboarding over Zoom/Teams sucks. So I built a way to whiteboard directly to your webcam - no screen sharing, no mouse-drawn gymnastics, no installs on a locked-down work computer. Draw. Explain. Iterate. Demo 👇 github.com/Techryptic/jes…

@infoxiao I still use mine, light weight and coding with it is seamless

justice for google glass

Got accepted to the @xai hackathon. I’ll be in SF through the 9th — open to meeting up if you’re around. DM me!

@temha @film @coffee I hope not 😁

X, aktif olmayan kullanıcı adlarını satmak için Handle Marketplace adında yeni bir pazar yeri kuruyor. “@film”, “@tech”, “@coffee” gibi nadir adlar artık açık artırmaya çıkacak. Sosyal medyada isim bile artık bir varlık sınıfı. 💰

FLOPs for All: GPUs & AI Compute in 3 Minutes! 🚀

Here’s one of my advanced vibe coding tricks that automates the process of compiling and debugging issues in native iOS app development. --- Compile the iOS app using Xcode with this command. Wait for the response, and if errors or warnings occur, fix them automatically and rerun the command. Repeat until no errors or warnings remain. xcodebuild -project [project_name].xcodeproj -scheme [scheme] -configuration Debug clean build CODE_SIGN_IDENTITY="" CODE_SIGNING_REQUIRED=NO 2>&1 | grep -E 'warning:|error:|❌|⚠️|BUILD FAILED|following build commands failed:' If errors occur, fix them automatically and rerun the command until the build succeeds without issues.

@TheKyaniL Both if I can, if not ill like to add in .app in the mix

do you prefer .com or .ai for an AI wrapper?

Huawei’s Ascend 910B is a case study in geopolitics and tech. It’s ~20% better than NVIDIA’s A100 in some inference tasks, but training large models exposes CUDA’s edge. With the 920 coming, China’s betting big on domestic AI hardware. Global markets, take note. #AI #TechWar

I deployed 2 iOS apps recently. One cost $350 in Claude 3.7. The other? $25 with 3.5. Same outcome. 3.5 wins, easy. If you know, you know. 🔥📱

@AnthonySecurity Glad folks are seeing the cert bubble popped

Protip: don’t bother with TCM, it’s not going to help you pass screening at any serious place

4/ Most exciting part: o3 demonstrates emergent reasoning strategies - discovering its own ways to validate solutions without being explicitly trained to do so.. 🔍

3/ Trade-offs are real: o3 needs less sampling (1k vs 10k solutions) but likely requires massive compute resources. But this points to quality over quantity in reasoning. 💭

🔥 OpenAI's new paper shows fascinating evolution in competitive programming: from hand-crafted strategies to emergent reasoning. Their o3 model hits 99.8th percentile on CodeForces through pure learning. Paper: arxiv.org/pdf/2502.06807