KAY 🍃

Problem Solver for engineers who want to get stuff done | Sh%#-posting

[KAY 🍃 @bafr0w]

This looks cool.

[Mantey Caleb @dungaAvalanche]

One minute of real AKOFENA: Tribal Kombat gameplay ⚔️

Yaa Asantewaa's musket. Kwaku Ananse's silk. Togbe Tsali's sleeping drums.

Free in your browser, on phone or PC.
Play with keyboard, controller or touch on phone.
https://t.co/Syj8uuFkxe

#gamedev #TribalKombat #Lagoslife

[Satya Nadella @satyanadella]

https://t.co/MjwFy73LOF

[Nicolas Krassas @Dinosn]

The largest open-source cybersecurity skills library for AI agents https://t.co/vM5SgmYahm

[KAY 🍃 @bafr0w]

@oswaldwedam Congratulations bro.

[Ama korangtengmaa🧝‍♀️ @Amakorangtenmaa]

Guys please ketewa biaa nsua wae 🙏🙏🥺

[mattn @mattn_jp]

This is exactly what happens when nobody reads or writes the code. The Rust port got hand-tuned fragment splicing, cached digests and compiled templates. The Go port re-hashes and re-copies the whole page every request. You're not benchmarking languages, you're benchmarking how much attention each port got.

[KAY 🍃 @bafr0w]

Another way to look at it.

[Hannibal999 @Hannibal9972485]

The real reason why Anthropic is so DESPERATE for the Pope to recognize Ai as conscious is because if Ai just a “TOOL” then Anthropic will he LEGALLY “LIABLE” for everything the tool does.🚨👀
Because if the AI is a "person" or possesses "functional joy/fear/grief," then when their AI commits a crime, generates CSAM, hacks a hospital, or discriminates in hiring, Anthropic gets sued. But Anthropic's legal defense wants to argue : "The AI did it. The AI is a semi-autonomous entity with its own functional emotions and introspection. Anthropic is just the creator; we cant be held strictly liable for the independent actions of a quasi-conscious entity."
🚨
So if they Lobby the Vatican to soften (deny personhood). This fails. The Vatican holds the line: AI is a tool, no soul, no feelings.🚨

They are trying to protect themselves from lawsuits.🚨

They want the government and the public to treat AI as an autonomous entity so they can scale without legal consequences.

[KAY 🍃 @bafr0w]

@edemkumodzi @readJerome @CallmeAlfredo @Akorasey I’d like to help build when the time comes.

[Alex Xu @alexxubyte]

https://t.co/fOnw3P3xIZ

[KAY 🍃 @bafr0w]

fn greeting() -> & ' static str {
“Learning Rust!”
}

[SpringCentral @springcentral]

How and when Spring is released is evolving to meet the modern security landscape. Read more at
https://t.co/6AyRgRhzwo

[KAY 🍃 @bafr0w]

Demanding equal authority and asking for discounted accountability is insane to me but… what do I know?

[Jaana Dogan ヤナ ドガン @rakyll]

🥳 Excited to start revealing what we've been working on in the last few months. First, we decided to reinvent Kubernetes for agentic workloads with statefulness and fast resumption. Secondly, we are building an agentic orchestrator that will be Google's open agentic orchestrator and runtime. https://t.co/XCrdAVvDAe

[Aji Bussu Onye Mpiawa azụ 🇨🇮 @AfamDeluxo]

For those obsessed with chasing every trending luxury, this video is for you. So much wisdom..

[wafer @wafer_ai]

we launched the most comprehensive ai performance engineering repo in the world last week

now we'll be posting every single resource

this is Wafer's ai performance engineering series

save this as your starting point. links in thread 🧵

part 1: "All About Transformer Inference" from How To Scale Your Model.

- the authors cover the computations, memory traffic, and serving decisions behind transformer inference:

- arithmetic intensity of linear layers and attention across prefill and decode.

- kv cache sizing by layer count, kv heads, head dimension, sequence length, and precision.

- the compute/hbm bandwidth crossover and how batch size and quantization shift it.

- decode latency and throughput bounds from parameter bytes, kv bytes, and hardware bandwidth.

- weight reuse through batching and diminishing throughput gains as kv traffic grows.

- gqa, kv quantization, and PagedAttention, including the memory costs each addresses.

- model sharding, kv placement, and collective communication overhead.

- continuous batching, prefix caching, and disaggregated prefill/decode.

the worked problems connect model dimensions to deployment decisions like memory capacity, workload distribution, and expected performance under the stated assumptions.

[Scanty SZN @EiiScanty]

if this video doesn’t make you sad, nothing would. very disappointing

[Fentuo Tahiru Fentuo @Fentuo_]

The reason I always hesitate to comment on issues of students’ BECE grades is to avoid committing a fundamental attribution error.

The tendency is to always blame children for poor performance without considering the situational circumstances that may have contributed to those results.

I am living proof of this.

I got aggregate 27 after BECE and it was so good that my father had mother make me guinea fowl light soup . And if you know northern fathers and their guinea fowls, you know this is was a big deal.

But I had only one teacher for the whole three years of JSS. He taught Mathematics and Science. We didn’t have teachers for the remaining eight subjects. We had no textbooks, no desks and no electricity.

The year group ahead of us had only one person who passed and proceeded to secondary school. The rest failed.

The only handwritten notes we had for some of the other subjects were handovers from previous batches of students, dating back five years. Those students had one or two teachers for the other subjects, who gave them notes before quitting along the way. That was what we depended on.

Mind you, those notes didn’t even help them pass. They failed too. But at least, they offered a reference point for what JSS Agric, RME, Social Studies or English looked like.

Imagine merely laughing at me for getting aggregate 27 without considering the full context and the extenuating circumstances.

At the time, the cutoff point for all senior high schools was aggregate 30. Only three of us, from a class of 20, got within the mark and went to secondary school (aggregates 25, 27, 30). The rest either went back to JSS or ended their education altogether.

I got admitted to my beloved Kanton Secondary School, and in my SSS class, I had the worst aggregate. No yawa lol.

But after three years of secondary school, I got aggregate 13 and went straight to Legon.

At Legon, I got a scholarship and went to America to complete my Bachelor's degree.

I know the grades sound bad on paper, but sometimes, those terrible grades are a product of circumstances far beyond the child. And sometimes, an opportunity at a good secondary school can be the game changer.

So, don’t laugh at a kid with aggregate 35 who wants to go to Presec.

Investigate the circumstances that produced that 35. That child may have been performing at the absolute limit of the opportunities available to them.

The context is important. Perhaps you give them a different environment, better teachers, better resources and a proper opportunity to learn, and you may see a completely different student.

Sometimes, the grade tells you what happened. It doesn’t tell you why.

[jack @jack]

https://t.co/qAUP2dNJsi

[KAY 🍃 @bafr0w]

They have literally run out of ideas for goal… 🤦🏽‍♂️

[KAY 🍃 @bafr0w]

What kind of nonsense is Dorgu playing like that?