Ra

I'm just derping around, spreading joy

[Ra @pivonroll]

@milosdj Ko jos gleda TV? Tehnologija 20. veka.

[Ra @pivonroll]

@PythonPr Then why numpy is used so much?
Go back to school kid.

[Ra @pivonroll]

@sri9s Nothing.

[Ra @pivonroll]

@hthieblot Forums with authentic people.

[Ra @pivonroll]

@GitHubCopilot Your latest update broke BYOK for Anthropic models on JetBrains IDE.
Error: Model 'claude-opus-4-6' not found on provider at https://t.co/cLFdZIp5Th (HTTP 404).
Fix that.

[Ra @pivonroll]

@Milan_Baja Novac devalvira. Tih 154k za 20 godina ima istu kupovnu moc (mozda malcice vecu) od tih 80k sada. Jos bolja varijanta je da vratis tipa 20-40k za 5 godina ako mozes i onda vracas 40-60k na 15 godina, rata tipa 200-300 evra. Cik nadji toliku kiriju da stoji 15 godina.

[Ra @pivonroll]

Zna li ko zasto Wolt ne dostavlja? Štrajk dostavljaca?

[Ra @pivonroll]

AI vibe coding is the new brain rot.

[Ra @pivonroll]

@CinematicEye_ Edited out

[Ra @pivonroll]

@d_pavlovic Zavrsio sam matematicki faks. Nikada ovako nesto ne bih davao ucenicima srednje. Cisto izivljavanje.

[luthira @luthiraabeykoon]

We implemented @karpathy 's MicroGPT fully on FPGA fabric.

No GPU.
No PyTorch.
No CPU inference loop.

Just a transformer burned into hardware, generating 50,000+ tokens/sec.

The model is small, but the idea is not: inference does not have to live only in software 👇

[Ra @pivonroll]

Get ready kids. New job roles: Maintainer of AI generated code (crap).
Most of the generated AI code is on the junior level and lower. I am talking about the backbones of the system. Not just some random function here and there.

[Ra @pivonroll]

Coding agents are like junior developers. I ask them to do one thing. They nod, go there way and come back with something which works in 60-70% of cases.

[Jordi Ribas @JordiRib1]

I’m pleased to share that our search team has open sourced an embedding model called Harrier that is currently ranking #1 on the multilingual MTEB-v2 benchmark leaderboard.

Harrier delivers SOTA performance on retrieval quality, semantic matching, and contextual analysis across workloads, supporting more than 100 languages and handles long inputs up to 32K. It is built for the next generation semantic search for Bing and our web grounding (RAG) service for AI agents, which already powers nearly every major AI chatbot today.

As you can see in the leadership board, our Harrier model is currently ahead of other excellent models based on Gemini, Gemma, Llama, Qwen, and more. I’m grateful for the hard work of our team to get to this top ranking, and I’m excited to see all the healthy competition in the space, which should ultimately lead to more innovations that will benefit everyone.

Learn more: https://t.co/tvEvCzk7Mf

[Ra @pivonroll]

@notgrubles I am betting in it npm and python dependencies.

[Ra @pivonroll]

@kareem_carr If code is bad Claude cannot help.

[Noah @NoahKingJr]

Vibe coders debugging an app they built with Claude Code:

[Ra @pivonroll]

@abhitwt And slow UI

[Ra @pivonroll]

@FFmpeg Is this a joke???

[Ra @pivonroll]

@ftn_se_budi Samo bez kulta licnosti.