Ae Nense retweeted
Why the fuck are Indians hated so much? What have they done to deserve this level of hatred?
Have Indians promoted terrorism? Have they colonized your countries? Enslaved your people? Stolen your homes? Taken away your families? Kidnapped your children? What exactly have millions of ordinary Indians done to deserve this much racism and hostility?
Is it because they're educated, intelligent, ambitious, hardworking, and competing for the same jobs as you? Since when did intelligence and ambition become crimes? Since when did working hard to build a better life become a reason to be hated?
If your problem is job displacement, then why aren't you directing the same anger toward artificial intelligence?
Why not question AI? Why not question ChatGPT, Claude, Grok, Anthropic, OpenAI?
AI has the potential to displace far more jobs than Indian immigrants ever could. Yet somehow, the Indian engineer, the Indian student, the Indian doctor, and the Indian worker become your enemies.
Why? Because it's easier to target a human being with a different skin colour or accent than to confront the actual economic forces changing your world?
And understand one thing clearly.
If an Indian refuses to engage with your racism, if they ignore your insults, if they choose not to respond to your hatred, it doesn't mean they lack a spine.
It means your hatred isn't worth their time.
Indians don't owe you an apology for being educated, ambitious, hardworking, successful, or simply existing.
Being Indian is not a crime. Being successful is not a crime. And your insecurity is not a justification for racism.
This thing is so load it can burst your eardrums if you are too close.
Also, the thruster tank gets emptiesd in seconds, and it won't ever be street-legal anyway.
Pure marketing stunt to protect Elon's brand value as some sci-fi genius from Mars.
Ae Nense retweeted
The Modi luck strikes yet again.
Barely 48 hours into Elon’s meltdown, and he single-handedly convinced the entire country why handing critical telecom infrastructure to an unpredictable billionaire is a massive security risk.
Modi looks like a visionary without lifting a single finger. 😃😃😃
BREAKING: Zelensky says he has been betrayed by Trump, that Elon Musk and Trump deliberately blocked Starlink in Ukraine overnight and that people have been killed as a result. "I am sure they knew what they did, absolutely. Because it was 1am at night, Starlink was blocked by SpaceX and the United States," he tells Piers Morgan.
"This decision will bring a lot of human losses. Of course, people will die." He said interceptors that rely on Starlink did not work, and that "this is not the right side of history."
Asked if he is angry with Musk, Zelensky said yes, calling it a big mistake.
"Now Putin is laughing because, I mean this is for him, simple victory," he says on the Trump-Russia diesel deal.
Ae Nense retweeted
For over a year now, anti-India racism on X has enjoyed freakish reach, engagement/payouts. Every Indian here has watched it happen. Racism against Indians literally incentivised, the X algorithm pours monetisation fuel on it.
Now a mass shooting plan by a white terrorist at a Hindu temple in Chicago has been foiled.
Hate spills from SM into real life, the people tweaking the algorithm will have blood on their hands.
Ae Nense retweeted
Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel.
- Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all.
- Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract.
- Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code.
- And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while.
The funnel is driven by a combination of factors:
1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel.
2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel.
3. Access. Free-tier, paid-tier, internal.
So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.
Judging by my tl there is a growing gap in understanding of AI capability.
The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code.
But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along.
So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions.
TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.
Ae Nense retweeted
Proudly designed & manufactured in India 🇮🇳
Ae Nense retweeted
You are being too clever by half, dear @elonmusk, and you know it.
Saying "11,256" villages sounds a lot. But it is a way of lying with numbers.
What if the handle had said, "1.7% of India's villages have no 4G coverage"?
That wouldn't fit the narrative, right?
And it only gets worse.
The post deliberately chooses "4G".
Why?
Do those villages not have access to 2G and 3G internet? Perhaps you know. The world's richest man who wants so desperately to enter the Indian market with satellite internet coverage should know, right?
You, who have talked about the 'Deep State' would know the perils that satellite can pose to national security. The terrorist attacks on Mumbai, better known as the "26/11 attacks", were facilitated, in part, by the use of satellite phones. The terrorists' handlers, sitting in Karachi, communicated with the terrorists using satellite phones and ordered them to kill as many people as they could, including Jews.
As for "oligarchs", well, you mean Mukesh Ambani (his mobile and internet company, Jio), yet hesitate to even taking his name. You, who didn't think twice before posting Trump's alleged links with convicted pedophile Jeffrey Epstein!
I think that will go down well with most Indians, to see the world's richest man in fear of an Indian🙂
Which brings us to the elephant in the room - racism.
It would take a blind man on Pluto to not have seen the explosion of racist screeds and shitposting and ragebait content against Indians over the last year-and-a half on your platform, X.
Perhaps it got your platform more engagement. Perhaps this is what the MAGA gang wanted and you obliged to get back in Trump's good books.
Perhaps you didn't care about racism against Indians.
Perhaps you saw it as a bargaining chip in your negotiations with the Indian govt.
In either case, you don't have a lot of sympathy in India because of racism on your platform, Elon.
Please stop pretending that there isn't racism on X against Indians, or that you don't know about it.
Coming back to broadband coverage.
The areas that do not have adequate 4G and 5G coverage are Pakistan Occupied Kashmir, parts of Rajasthan that border Pakistan and which are sparsely populated, the Kutch region (also sparsely populated, what with the world's largest salt pans), Arunachal Pradesh, tracts that are either heavily forested or till recently had suffered from Naxal terrorism.
So, what exactly is your target market?
Offer satellite coverage in Arunachal Pradesh and China will cut off your businesses at the scrotum before you can say Tesla.
Or did you plan to make a killing selling satellite coverage at $50/month in Bihar and Jharkhand, where 4G and 5G plans cost $3/month?
Get honest. Get real.
xitter.cf/AbhinavAgarwal/status/…
myvi.in/vicoverage/
ookla.com/articles/india-mob…
Ae Nense retweeted
>see "Adobe is done" all over timeline
>wow, entire suit rebuilt
>"Every Crafting App is written from scratch in Rust and held to the same rules."
>"native, not wrapped"
>Adobe must actually be over
>download one app
>5gb RAM idle, 100% CPU, not responding
>look inside
>Tauri
They said "pls no electron or tauri", AI did it anyway, but sloppers don't look at the code or use their own stuff.
I think Adobe will be fine
Readers added context they thought people might want to know
The Crafting Apps shown (e.g. PhotoCraft) use native egui/eframe, not Tauri. The separate ArtCraft AI app uses Tauri. Poster confirmed checking ArtCraft.
github.com/storytold/phot…
getartcraft.com/apps
Ae Nense retweeted
I'm going to spill all of my secrets here and use this as a public teaching moment on how to make viral videos about bad public infra.
One of the most important things is your thumbnail (on X this is the first few frames of your video). This should feel organic, not heavily edited/modified. Just you, the background, and 3-4 words of superimposed text works well in my experience. Three words is the sweet spot. Four is the maximum. Five and you are expecting people to read too much as they doomscroll, which will hurt CTR. I find that bright yellow text helps to make it pop. You could also use bright green, orange, red, or any other eye-catching colour. White text (in my opinion) tends to not pop unless it's encapsulated within a colourful box/container of some kind.
Notice how, for the image on the left (your thumbnail), there's nothing to actually convince me to watch the video in the first frame. We see you (handsome, but no virality factor I'm afraid), the background (nothing much to note there, it doesn't embody what the video is actually about), and some text: "Hi, I am right now on Silver Country Road which is an" - this text doesn't really tell me anything. If I don't live on Silver Country Road (which I don't) then this video will not feel instantly relevant to me.
In my 12th Main video, @Virendra0698 and I actually shot the intro scene at the end, after finishing the main shoot. My initial thumbnail concept was not strong. I only realised this as I began editing the video on my phone while I was still on 12th Main. We went back to the same spots we had shot earlier and I sat down in the asteroid crater, sat inside of the drain (it did not smell great), and leaned against the pile of rubble. I felt that these would be shocking visuals that quickly captured the essence of what the video was about. I also think the quick cutting between those initial shots captured viewer attention and satisfied people who have a short attention span.
Apart from the thumbnail, think about your opening hook. Instead of, "Hi, I am right now on..." you could start with "This might be one of the worst footpaths in Bengaluru" or "Footpaths in Bengaluru are bad, but this one takes the cake" - generalise the topic and make it feel relevant for a large target audience. Get straight to the point too, don't beat around the bush with introductory information or pleasantries. You want people to feel a visceral emotional response in the first 1 second of your video. This could be rage, disgust, shock, sadness, etc.
In the image on the right (my AI-generated thumbnail - normally you would not use AI for this ofc), the viewer has all of the info they need to make a decision on whether or not they want to watch this. The text very clearly mentions that this is about Bengaluru's Worst Footpath (the question mark is optional - you could remove it, or leave it if you want to spark some curiosity). We also SEE the footpath itself (visual proof) in the background, covered in trash - this helps us to realise that this is not a talking head video, it's an actual vlog-style exploration of the terrible footpath. And then there's you, the main character, who will be guiding us through the experience. Humans are always attracted to content with other humans, so this is an important ingredient as well.
Another trick is to make sure all of the core information is contained within a 1:1 aspect ratio. In your thumbnail (left) the subtitles are actually located outside of that 1:1 square, so if someone quote reposts your video, people may see a cropped preview that DOESN'T contain that critical text. QTs are a huge part of X success, so you should be optimising for that. This 1:1 safe zone is good hygiene across platforms too. Instagram also crops videos and thumbnails depending on how/where the viewer is encountering the content, and UI elements also block visibility on the top, bottom, and right side of the frame. Google Search also does an even more aggressive 16:9 crop, wherein vertical thumbnails become horizontal ones. If your video blows up, Google may be where many people actually discover your video.
I'd recommend that you post this video across X, YouTube, Instagram, and LinkedIn to maximise visibility. My 12 Main footpath video has more than 2M views on Instagram, whereas on X it only has 4 lakh views. Given that the goal of this video is to get the thing fixed, the more eyeballs the better. You can also send collab requests to prominent Bengaluru news/updates/drama accounts on Instagram, which can accelerate visibility if they accept your collab (which they will do if it starts to go viral).
I generally do not recommend tagging anyone on X, as this has been known to hurt reach (might have changed with X's new algo though). That said, in this case tagging me did work, so I guess this is an exception to that rule.
Keep videos under three mins. This allows you to post it as an Instagram reel and YouTube short without restriction.
You crushed it with the audio quality. I could hear every word. Adding subtitles is also a must. Great job on those. Just move them up so they're in the 1:1 safe zone. People often struggle with audio quality if they do not own a professional mic. Using a secondary phone as a mic was a smart move.
Lastly (and this part is optional but recommended) to really increase your chances of going viral, you should come up with your personal masala. Something that will cause your video to stick in people's brains for days or weeks. You can become the "_ Guy" using this masala. For me, it's my skin colour/nationality. I'm a white guy in India. In one of my early footpath videos, I navigated the footpath like I was doing parkour, and the headline became, "Canada Man Performs Parkour on Bengaluru Footpaths." This was a double dose of masala. For @baadalvirus he dressed up as an astronaut to spotlight the moon crater-like potholes in Bengaluru. Another time he got journalists to dress up as Squid Game characters and navigate a broken footpath like it was a deadly obstacle course. Residents of Cox Town held a "pothole pooja" to raise awareness. People may not remember your name or your face, but they will remember your masala.
Overall though, great video jstar. Please keep up the good work, and thank you for raising awareness. You're one of few people who has actually made an earnest attempt to highlight issues with an engaging, thorough video, rather than doing what most people do which is just snapping a pic or shooting a low-effort video without any subtitles and angrily tagging a bunch of politicians.
Got inspired by @caleb_friesen to make a video on the poor condition of Haralur Main Road
This stretch is right in the middle of Cambridge international school and National public school, and it pains me that this is the infrastructure we are able to offer to the future gen
Ae Nense retweeted
Replying to @notsky
Constantly throughout the months it is up for. Most of the things that went into this branch were separate PRs we reviewed and merged in.
Obsessive real world testing (and programmatic testing) helps as well, but it’s not enough. The nightly system helps a lot here, reducing blast radius of broken code and assumptions
Ae Nense retweeted
Crashing with a friend in Baiyappanahalli just beside Karnataka High Court's staff accommodations.
The footpath is a pedestrian obstacle course made of:
- Illegal construction encroachment
- Broken slab death pits
- Mountains of trash
Why do Bengaluru residents tolerate this?

























