Aarkey Narayan

I am RK, Indian 🇮🇳

[प्रेम शंकर चरडाना @PremSmeena49]

कंपीटिशन एग्जाम के फील्ड में नये आये लोगों को लगता है कि ‘ये तो मैं एक बार में ही कर दूंगा’ लेकिन कब ये ‘अगली बार तो जरूर करूंगा’ से, ‘अब नहीं होगा!’ बन जाता है, पता भी नहीं चलता।

[Blake Burge @blakeaburge]

Underrated life advice: Have more hobbies and fewer opinions. Learn an instrument. Plant a garden. Build something with your hands. Cook. Paint. Run. The happiest people I know spend less time debating life and more time actually living it.

[Prompter @PromptLLM]

Absolute banging advice from Claude

[bebo ☾ @smacklaced]

Average sink in a country where pursuing art as a career is frowned upon

[Netflix Anime @NetflixAnime]

Where it all began: East Blue.

Produced by WIT STUDIO, THE ONE PIECE sets sail in February 2027. All 7 episodes dropping at once, only on Netflix.

[Netflix Tudum @NetflixTudum]

FEBRUARY!!!!

[ONE PIECE.com(ワンピース) @OPcom_info]

5月5日はモンキー・D・ルフィの誕生日

HAPPY BIRTHDAY
MONKEY.D.LUFFY

#ルフィ誕生祭2026
#ONEPIECE

[Aarkey Narayan @aarkey_narayan]

@Poojakumar89308 To ab inhe reservation nhi milega ? They are now GC ?

[Miguel Conner @AeonByte]

Gnosticism is a niche ideology that has no impact on modern cultur--

[Y Combinator @ycombinator]

AI has stopped being a feature and started being the foundation.

We're excited about a new wave of startups rebuilding software, services, and silicon— and pushing AI into the physical world.

https://t.co/QCIz6DnQnN

[Aarkey Narayan @aarkey_narayan]

@Sachin6597 Ghar se chla jata yarr, itna hi harami jaalim baap tha to, nahi rhna tha sath me aur kya

[Aarkey Narayan @aarkey_narayan]

@prakashtmk8 @ComradeMalal Ekdm sahi baat, biharsharif me to Kiran sir, Mastermind, Pankaj sir, etc etc falana dhimkana, sara sir log apne yha aane wale 11-12th ki bacchiyon ko fasa ke rkhe hain. Ye main 2016-17 se dekh rha hu.

[Aarkey Narayan @aarkey_narayan]

@raavan_india Biharsharif, muzaffarpur

[Aarkey Narayan @aarkey_narayan]

@pixreload Avg Boring road kanya

[Aarkey Narayan @aarkey_narayan]

@WhateverVishal Meerut ?

[Aarkey Narayan @aarkey_narayan]

@gemsofbabus_ @amitkilhor kyu re

[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.

[DAN KOE @thedankoe]

https://t.co/xDEP4ad1tz

[Aravind @aravind]

All the cooks with Youtube channels, here's an idea:

More than a century before Europeans wrote their first "cookbook", India had a book written in Karnataka called Manasollasa with an entire section on recipes of the day.

It describes making Dosaka (Dosa), Idarika & Vataka (Idly & Vada), Kshiravataka (Dahi vada), and dozens of other dishes like fried fish, chicken, pakoras, sweets etc. It contains methods of preparation of Indian wines and liquors too.

And it uses truly Indian ingredients as there were no tomatoes, chilies, potatoes etc those days. The book was written in 11th century India and is the earliest surviving book containing Indian recipes.

So all the great cooks with YouTube channels can try this: Do your research and get hold of a translated copy of the recipes. And try to make some of them on your channel, introducing your audience to ancient Indian cooking and dishes.

Here's a fried fish recipe from 11th century India:

- Cut white fish into pieces and wash them well
- Cook them in tamarind juice
- Coat them well with wheat flour
- Fry in heated oil till brown
- Season with kalanamak, powdered cardamom and pepper.

[Your Best Version @YourPrimePath]

I fell in love with this quote:

"No matter your age, you'll always wish you started younger, but today is the youngest you'll ever be. So start today."