Balakrishnan

@Pranesh_Balaaji interesting bet

DMK will win comfortably. TVK will lose deposit in almost 200+ constituencies

@Pranesh_Balaaji can you back with data? and since when

Whoever wins the Kangeyam constituency decides the fate of Tamil Nadu elections. Yes, you read that right; every party that has won Kangeyam has gone on to win the state.

@DearS_o_n Pharma. Food industry. Schools. Banking system. Marriage. Taxation. Luxury. “Independent” media. “Non-governmental” organizations. Cities. Countries. Reality itself.

We all love talking about growth. Navin’s actually done it — built Tumbledry from scratch to 1500+ stores. Now he and Nikita are at it again with Duck Duck Baby, a kids brand out of Delhi NCR. Had them over at Roopac today, and honestly, just the stories were worth it. That gets even better when we get to build their packaging?! 🥹

@karpathy Drop the video on this, Andrej.. long time 😻

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.

what is this new thing with leaders of both sides- dmk and admk- draping towels with party flags in checks. 😍

Annamalai also same feeling.

You can still instantly tell when an app lacks the human touch. I can't commit my time to something if I see that the creator is not willing to commit theirs.

How many paper straws do we need to use to offset these emissions?

Why not AS Babu Sah just can’t make a better Saree box

And then Annapoorna - very slow to this game of scaling - is killing it with the craft

I did not know this. Daniel Kahneman, Nobel prize winner and author of Thinking Fast and Slow (which inspired my Artificial Intuition book), passed away with assisted suicide. I heard about this in a recent interview with Ray Kurzweil. Here is his final note:

The cold face at home since the conflict

@AmazingCreditC @mehulshahca in case it slips, there is just no fallback/ service. you can buy with this known downside

@bkrishn @mehulshahca What if I connect it to a TV stabiliser ?

I have to buy a new TV for our Living Room. Any suggestions as I do not see any innovation or anything new as a feature in this space?!

@AmazingCreditC @mehulshahca with an awareness that even a slight power fluctuation- during the regular software update (which you can’t opt out) —would corrupt the software. and the new models are non serviceable afterwards!

@mehulshahca No matter which TV you buy, make sure to get an Apple TV 4K. Most TV operating systems these days are quite laggy. Planning to get one soon myself. amzn.to/4aOdgT3

If you are a company registered in India, it’s starting to make more sense to buy from Apple store - with tax credits, it’s near same as in the US But what work you do with it is the broader and relevant Question

The single most power every business today has and can act is Claude code - unthinkable what you can build today vs what was possible even a year ago. We are stopping every other thing we do- to just code. The entire team - design, ops, sales

@ANI Look how sweet the Chennai boy speaking on his new dapper suite

Delhi | India AI Impact Summit : Sriram Krishnan, Senior White House Policy Advisor on Artificial Intelligence, says, "...We want to make sure that the world uses the American AI stack...We also want the world to use our AI model...We want all our allies, including India, to leverage our AI infrastructure."

Immortality is attained by having children, leaving the world in a better place than you found it, and pushing the human frontier. Not by drinking “longevity mix powder”.

கலைஞர் காப்பீடு, அம்மா உணவகம் வரிசையில்! But cool site!