OpenAI’s workloads run on Microsoft’s Azure AI cloud platform, on Oracle’s data-center infrastructure

🤔 (source)

lol

is this number evil?
codegolf

python solution:

lambda n:int(bin(n),13)%2

DeepSeek-R1
A competitor to gpt-o1
with MIT license

https://github.com/deepseek-ai/DeepSeek-R1

#llm

And it's funny how it shows the CoT and then censors the output xD

Turn any Git repository into a simple text digest of its codebase.

This is useful for feeding a codebase into any LLM.

https://gitingest.com/
repository: https://github.com/cyclotruc/gitingest

FunC, a high-level language, is used to program smart contracts on TON.
It's a domain-specific, C-like, statically typed language.

Here is a smart contract's source code for example.

FunC programs are compiled into Fift assembler code, which generates the corresponding bytecode for the TON Virtual Machine (TVM).

The bytecode (essentially a tree of cells, like any other data in the TON Blockchain) can then be used to create smart contracts on the blockchain or can be run on a local instance of the TVM.

- FunC Cookbook
- FunC docs
- Source code of TVM (C++)
- TON & Telegram Blockchain Сourse (you need to change the site language to English, by default it's set to Russian)
- TON paper (Author: Nikolai Durov)

Introducing Operator
Article, Comments

memos
An open-source, lightweight note-taking solution. The pain-less way to create your meaningful notes. Your Notes, Your Way

GitHub: usememos/memos

#go #react #self_host

g.co, Google's official URL shortcut, is compromised (as email)
Article, Comments

https://github.com/metacall/gsoc-2025

List of project ideas for contributors applying to the Google Summer of Code program in 2025 (GSoC 2025)

Survivorship bias

Diagram in which red dots stand for places where surviving planes were shot. This only tells you where planes can get shot and still come back to base. Survivorship bias: your only information is what has survived.

Survivorship bias or survival bias is the logical error of concentrating on entities that passed a selection process while overlooking those that did not. This can lead to incorrect conclusions because of incomplete data.

It's when we focus only on the people or things that made it through ("survived") some process while overlooking those that didn't, leading to incorrect conclusions.

Here's a real-life example that's easy to understand:

Imagine you're looking at successful entrepreneurs like Mark Zuckerberg or Bill Gates, who dropped out of college and became billionaires. You might think, "Hey, dropping out of college is the path to success!" This is survivorship bias because you're only looking at the "survivors" - the few dropouts who became extremely successful. You're not seeing the thousands of people who dropped out and struggled, or the many successful people who completed college.

Here's another everyday example:

Think of an old neighborhood with houses built in the 1900s. Someone might say, "They don't make houses like they used to. Look how strong these old houses are!" This is survivorship bias because they're only seeing the houses that survived for 100+ years. All the poorly built houses from that era have already been demolished and replaced - you can't see them anymore!

The key lesson is: to avoid survivorship bias, we need to look at *both* the successes *and* failures, not just the "survivors," to get the complete picture.

Kitty

Cross-platform, fast, feature-rich, GPU based terminal, written in Go (and C), with scripting framework in python.

https://github.com/kovidgoyal/kitty

#go #terminal #tui

51% attack

A 51% attack is an attack on a cryptocurrency blockchain by an entity or group that controls more than 50% of the network. If a party were to gain this much control of a network, it would have the power to alter the blockchain.

e.g. if there are 100 nodes inside of a network and you own at least 51 of them, then you can control the network (e.g. approving transactions which would be rejected in a normal situation)

This is a vulnerability that still has not got any universal solution yet.

- Source