Anoop Menon

Never make fun of Indian English. It's OUR accent. Only make fun of liberals with fake foreign accents. - Great ad by 'Axij' Bank.

@itsarnavb - isn't this the same thing you reported ages ago?

Imagine you're John Carmack you're 22 years old and you just wrote a 3D engine in assembly that runs at 35fps on a 486 Doom drops. Quake drops. Half the planet is playing your code. you're the reason GPUs exist. you're the reason your friend Jensen has a yacht today. then in 2009, you sell id Software. people call it betrayal. you call it "they made an offer I couldn't refuse." VR obsession. Oculus. Meta buys it for $2B. you're CTO. but Meta thinks you're a liability. your demos are "too intense." your emails are "too long." your focus on frame timing is "slowing us down." 2022. they push you out. not fired officially. just "restructured." the media writes "end of an era." some crypto bro calls you "washed up." silicon valley moves on. but you don't. you don't write a book. you don't start a podcast. you don't collect speaking fees. you go completely quiet. you take the money. you buy a warehouse in Texas. you hire 10 engineers. and you start coding. not games. not VR. AGI. two years. radio silence. no tweets. no conference talks. while everyone's debating ChatGPT, you're debugging CUDA kernels at 3AM, testing world models. then in 2025, Keen Technologies pivots hard. you're not "exploring" anymore. you're building it. here's what people get wrong: everyone calls it a comeback. a redemption arc. "revenge on Meta." it's none of that. you're a 54-year-old engineer who still codes 12 hours a day because you genuinely can't stop. most CTOs would have bought an island. most legends would have written memoirs. you just kept typing. the most dangerous person in any codebase is the one who goes quiet and never stops shipping commits. karma doesn't need to be real. but obsession is. welcome back, Carmack.

here’s the biology behind this: what it is, how it actually works, why it works at all, and why it might not. 🧵 what the cells actually are they start with adult donor skin or blood cells, reprogram them into induced pluripotent stem cells, essentially hitting an "undo" button on the cell's identity, and then coax those stem cells into becoming functional human neurons. think of iPSCs like a git repo you've reset to an empty commit. you can now differentiate them into almost any cell type. neurons are just one option. those neurons are then grown directly onto a silicon chip with electrodes, inside a self-contained unit that includes life support, nutrient baths, temperature control, keeping cells alive for up to six months. the chip reads their electrical spikes. it also sends spikes back in. it's a closed loop. the cells are both the processor and the learner. how a neuron actually works a transistor is binary. it's on or it's off. a neuron is nothing like that. a neuron is a biological cable, a long branching structure with thousands of incoming connections (dendrites) and one output wire (the axon). it receives electrical signals from thousands of neighboring cells simultaneously, integrates them, and if the total voltage crosses a threshold, it fires, generating an action potential, a sharp spike of electrical charge that travels down the axon to the next cell. the spike is always the same size. what carries information is the timing and rate of spikes, not their magnitude. it's frequency modulation, not amplitude. the connection between two neurons is called a synapse. synapses aren't static. they physically strengthen when two connected neurons fire together repeatedly and weaken when they stop. this is Hebbian plasticity, "neurons that fire together, wire together." in deep learning terms: the weights are the hardware, and the hardware edits itself in real time based on usage. there is no separate training loop. every moment of activity is simultaneously inference and learning.

I'm so excited to introduce this! We've worked on a million different moving parts to produce this. I'm fairly confident it's the best multimodal model that exists, period -- and it's not too shabby at pushing back the LIMITs of retrieval either...

Every spring, Bengaluru turns pink. 🌸 These beautiful trumpet flowers have become part of the city’s identity. But very few people know the strange story behind them. The man who helped shape this blooming landscape… was once arrested by the British and sent to an internment camp as an “enemy of the state.” How did the man who made Bengaluru bloom end up in prison? Watch the story. Missing Chapter | Episode 1 — with the very talented Sathvika Kanala.

@Paperclip_In Thank you @Paperclip_In and Sathvika

bitterlesson-pilled folks here be like: .. i tried so hard and got so far.. but in the end, it doesn't even matter

A shout out to lovely folks @realpython. I have been a happy customer for over two years now. Due to some reason, I missed the renewal email but did not want to continue the sub. The lovely folks there responded promptly. Good to know such amazing businesses still exist!

@Soulthym @cmuratori @tsoding Thank you!

@nitroargo @cmuratori @tsoding I dont know of any other, but look for lsp enabled color highlighting systems where the lsp can raise "unreachable code" diagnostics or equivalent

Very often I want to temporarily disable a piece of code. I comment it out, but then I'm faced with a problem that since the code is never compiled it gets "stale". Some functions it uses may have changed and it is never type checked. So the next time I enable it, it doesn't compile and I spent a lot of time fixing it. The solution I came up with so far is to "comment out" the code with the runtime `if (0)`. The code will never be executed, the optimizer will very like eliminate the code entirely, but before doing so the compiler will type check it, and will force me to fix it on the spot.

@Soulthym @cmuratori @tsoding Would you have any more examples of such tooling? I am building out an app for CL language on IBM i that has this as its fundamental premise.

@cmuratori @tsoding Some languages do that already I think? Or at least some tooling does. I remember python with pyright in neovim commenting out disabled lines in some obvious cases (like `if False: ...`)

@can - help? Or maybe help me help you

@zebriez The Awk Programming Language. Starts with examples right from Page 1. dn790008.ca.archive.org/0/items/pdfy-M…

what is the best technical primer on a topic you've ever read?

@_svs_ for more context - x.com/ka_grieco/stat…

Something tells me this is psyops. US stockpiles are from here till eternity I imagine.

@_svs_ You should give this a try at least - simonwillison.net/2024/Jun/16/ji…

just wrote a claude skill called browse which strips all css cruft from an html page and now my agent can use my whole site. the thing is that html is actually xhtml. it's just xml. but because y'all tailwind pilled the html is basically unusable by anything with any intelligence. html also has all the information required to take the next action. it has links and forms. your json api doesn't spell this out. there have been some attempts like hateoas but they are too clunky. one of the things that rails pushes you towards is to not have a separate json api. the html and json views are just different formats of the same controller action. So you end up with a site that when given an api token, by default, exposes itself as an html api as well. and now i don't need to rewrite anything to be in json because everything just works. my agent now has token efficient access to my entire website, basically for free. you don't need chrome webmcp or whatever they announced the other day.

An IAS officer in Chhattisgarh FIXED a maternal health CRISIS that the government couldn't solve with a recipe older than modern medicine. She did it with a ladoo! Yes, you read that right 🤯 Okay so here’s what happened: Koriya district had one of the worst maternal health records in the state: → High-risk pregnancies → Underweight babies → Mothers going into labour severely anaemic. Simply because pregnant women weren't getting enough nutritious food. So District Collector Chandan Tripathi did something no consultant would pitch. She turned a grandmother’s ragi modak into a structured maternal health system. Here’s what they did differently: → Created iron-rich ragi modak ladoos (dietician approved) → Gave 2 ladoos daily to every pregnant woman → Added iron supplementation from the 5th month → Paired each woman with a “Poshan Sangwari” to ensure she actually consumed them They called it the Koriya Modak Ladoo programme. The most brilliant part about this is that they didn't hire outsiders to make the ladoos. The same women it was meant to help now make the ladoos, earning ₹10,000–12,000 per month. And look at the results now: ✅ 57% reduction in low birth weight cases. ✅ 362/398 underweight mothers gained healthy weight ✅ 3,00,000+ ladoos distributed so far. And all of this was possible not because of a ₹100 crore government tender but because of trust in community knowledge and the will to execute it properly. Sometimes the most powerful solutions aren’t expensive.

Nominative determinism strikes again: I met someone named Ruby who works on lasers.

@nikunjtaneja_ I bought STAY perfumes based on @shantanugoel 's recommendation. The magnetic clasp of the cap snaps on beautifully and is a nice design touch that probably is available on much higher priced products, if at all

i'm hiring Graphic Designer - Performance & Content full-time, in-person in Gurugram comp: 30-50k/month, will double this in 6 months 2-3 years of exp creating performance creatives & social media content at a brand or an agency email careers [at] stayparfums [dot] com

The shock was so profound that the Maharaja of Cochin reportedly had to intervene to control the proceedings. Some stories suggest the trial was halted when it became apparent that the next name on Thatri’s list might have been the king himself. Checkmate. 21/21

A new ad that managed to make me teary-eyed! Watch it first before reading any further, to avoid spoilers. 1/2