@katpadi.bsky.social

i write code 〽️
she/her 〽️
lakers 〽️
coffee 〽️

[@katpadi.bsky.social @katpadi]

Anunoby!!!!!!!!!!!!!!!!!

[���ィリピン住みます芸人 田中 @TANAKApinas]

フィリピン人女性、出勤前でも髪の毛がびちょびちょです。

フィリピンに移住して10年

最初の頃は、文化の違い全部に驚いてました。

でも10年も住んでると、それが当たり前になってきて、最近はあまり違和感を感じなくなってきています。

ただ、ボラカイ島に引っ越してきて1ヶ月

久々に、

「やっぱこれ変な文化だな…」

と思った事があります。

それは、街中のマッサ��ジ呼び込みのお姉さん達

なぜか全員、髪の毛がびちょびちょ

しかも1人じゃない

本当に全員レベルで濡れてる。

最初は「海入ってたのかな？」と思ってたけど、マニラに住んでた頃も、出勤前の女性達の髪がびちょびちょだった事を思い出しました。

フィリピンのローカル家庭って、ドライヤーやヘアアイロンが無い家も結構多いんです。

乾かさなくても、この暑さなら自然に乾く

だから気にしない文化なのかもしれません。

10年住んでても、たまにここ日本じゃないなって気付かされます。

ちなみに今では僕も、朝びちょびちょの髪のまま平気で外出るようになりました。

完全にフィリピンに染まってます。

海外住んでる人、今でも不思議に感じる文化の違いありますか？

#海外生活

[Andrej Karpathy @karpathy]

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest:
I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE:
I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A:
Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output:
Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting:
I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools:
I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations:
As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

[@katpadi.bsky.social @katpadi]

🤣🤣🤣

[Sam Amick @sam_amick]

There’s a woman courtside here in San Antonio who brought a mini Oscar trophy to present to SGA whenever he hits the deck. More sights/sounds/analysis on the @TheAthletic live blog here.

https://t.co/E8oH0QM43k

[@katpadi.bsky.social @katpadi]

Azzi Fudd unlocked!

[Women’s Hoops Network @WomensHoops_USA]

AZZI FUDD TODAY 🔥

• 24 POINTS
• 3 ASSISTS
• 3 STEALS
• 2 BLOCKS
• 6/12 3PM
• 9/15 FG

[禿道道🐟 @dearemon]

My Apple Car.

[Cursor @cursor_ai]

We’re introducing the Cursor SDK so you can build agents with the same runtime, harness, and models that power Cursor.

Run agents from CI/CD pipelines, create automations for end-to-end workflows, or embed agents directly inside your products.

[@katpadi.bsky.social @katpadi]

hello 2am, we back!

[@katpadi.bsky.social @katpadi]

*time travel noises*

[@katpadi.bsky.social @katpadi]

About to go back in time...

[@katpadi.bsky.social @katpadi]

🤯

[Claude @claudeai]

You can now enable Claude to use your computer to complete tasks.

It opens your apps, navigates your browser, fills in spreadsheets—anything you'd do sitting at your desk.

Research preview in Claude Cowork and Claude Code, macOS only.

[Lee Robinson @leerob]

Yep, Composer 2 started from an open-source base! We will do full pretraining in the future.

Only ~1/4 of the compute spent on the final model came from the base, the rest is from our training. This is why evals are very different.

And yes, we are following the license through our inference partner terms.

[@katpadi.bsky.social @katpadi]

@TalktoBPI Can't register my app after your update. It can't send an SMS.
Please fix asap.

[@katpadi.bsky.social @katpadi]

As Gloria would say...

It's a doggy dog world out there.

[@katpadi.bsky.social @katpadi]

wow last day of 2025 🤯

[@katpadi.bsky.social @katpadi]

@grabph But I've been using this for the past 2 years?

[@katpadi.bsky.social @katpadi]

@grabph Grab Pabili says unavaialble selected location for anything.

[@katpadi.bsky.social @katpadi]

Fail whale IMY

[Pankaj @the2ndfloorguy]

story behind "why netflix built https://t.co/YDCurkt2BM" is brilliant.

so, netflix had a massive fight with ISPs around 2014-2016. ISPs were slowing down netflix on purpose. they wanted more money from netflix

customers got bad streaming. but ISPs just blamed netflix.

netflix had to pay comcast, verizon, at&t and time warner for direct connections to their networks.

but in 2016, they launched fast dot com, clever part - It's not testing your general internet speed. It's testing your speed to netflix's servers specifically. so when someone complained about buffering, netflix could say "run fast dot com." If it's slow, the ISP is the bottleneck.

suddenly millions of people had a tool to prove their ISP was the problem
ISPs couldn't hide anymore.

netflix positioned themselves as the transparent good guys fighting for customers while ISPs looked like greedy monopolies

they solved a pr problem and a customer service problem with one simple website

I guess, that's how you win a corporate war