
AI partner for companies | IITM '22 Prev: AI @alma_inc, Ads @JioHotstar, Unacademy https://nitter.gq/t.co/uI46OjOm5V
Talk to me ➡️
Joined March 2016
- Tweets2.5K
- Following1.2K
- Followers594
- Likes11.3K
Replying to @amrnth0
Btw, if your company is devising it's AI strategy, I'm happy to help. Contact me:
cal.com/amrnth/20min
A friend DM'd me saying he's apprehensive of Deepi building Temple because he doesn't have the expertise in healthcare. This may be a reasonable argument for someone who hasn't worked on large projects, or who isn't an entrepreneur. You can hire expertise!!! Temple literally has cracked scientists in its team who specialise in the biology of the brain. Also, a take most people are yet to internalise -- expertise will be available on tap in the future. It is available even now to those who have the money to pay for it(AI). What will matter in the future is grit, perseverance and the agency to bring ideas into reality.
I don't get the Deepinder Goyal hate. I can't help but draw a parallel to Elon who gets hate from the Reddit-like crowd inspite of having so many achievements!
Deepi literally built a $35B company from scratch and inspired so many of new age Indian entrepreneurs (myself included).
Temple is a such an innovative deeptech product. Just read the specifications in the PDF Deepi posted. We can count on fingers how many such revolutionary technologies have been built in India.
Deepi also runs LAT Aerospace, which is trying to build India's first indegenous gas turbine engine for aircraft.
15
1,422
I don't get the Deepinder Goyal hate. I can't help but draw a parallel to Elon who gets hate from the Reddit-like crowd inspite of having so many achievements!
Deepi literally built a $35B company from scratch and inspired so many of new age Indian entrepreneurs (myself included).
Temple is a such an innovative deeptech product. Just read the specifications in the PDF Deepi posted. We can count on fingers how many such revolutionary technologies have been built in India.
Deepi also runs LAT Aerospace, which is trying to build India's first indegenous gas turbine engine for aircraft.
Replying to @DrSandipRoy
Shifting the goalpost farther away every few months is how you build large companies.
Went to the HQ of one of the top 3 ball valve manufacturers in India (north of ₹300cr rev.) to help ease their repetitive ops. Classic problem for AI -- specifications for actuators, accessories scattered across hundreds of PDFs, tribal knowledge retained within tens of employees who've been working there for ages, messy ERP, CRM software and so on.
I ended up advising them not to invest in automating some things as the cost of paying for AI inference would be much higher than if they'd hire an extra employee! Not to mention the cost of maintainence, over time, of overengineered solutions which just give small wins.
Other companies were pitching them to do everything for them. I could have taken their money and said I can do it all, but my conscience advised me against it.
Just Ask
Replying to @rikiyatakehi
5/6 Can an LLM improve another LLM?
We simply asked Astra to make Qwen3.5-0.8B better at math.
Astra saw Qwen get stuck in reasoning loops, which is a known issue that Qwen’s own model card warns about (see their official model card). With a budget of 50 forward passes on practice problems, it located the cause and edited it out. This used no gradients or training.
Fixing these loops by editing just 0.016% of weights boosted other capabilities beyond math too!
1
229
I reluctantly signed up to give a talk a month ago. I'd never spoken in front of many people. I have had a bad taste from trying and failing miserably to play guitar in front of people, one time, back in college. But this time, I somehow mustered the courage to speak without feeling small. And after the talk finished, and it'd gone well, I felt a rush of happiness I hadn't felt in a long time. I felt big, full of confidence, potent.
A week ago, a friend threw a party and I had a pint of beer after my friends pushed me. It had been ages since I drank. Let me tell you, I had a great time afterwards! I am usually not the life-of-the-party guy, but here I was cracking jokes, making everyone laugh and have a great time. I didn't regret drinking that beer.
So, trying to do things which you don't know you're good at would cause reward prediction error in your brain, and may lead to you enjoying it and having a blast.
Dopamine does not spike highest for guaranteed rewards. It spikes highest at reward prediction error. If you want to have fun in life, you should stop avoiding risk and uncertainty.
10
475
Forbes 30u30 strikes again😭
ED Bengaluru has conducted searches under FEMA on 27.10.2026 at 3 locations in Bengaluru & Madhya Pradesh over unauthorized cross border financial transactions using cryptocurrency involving more than ₹5,000 Crore (Approx.). The probe exposed use of shell entities in tax haven countries, massive OTC deals and illegal foreign payouts via platforms like Binance and Bybit in violation of FEMA provisions.
3
162
Since Von Neumann probes can take millions of years to move from one celestial object to another, clearly the human(s) inside these probes would have to live equally longer. This can only happen if we would have literally solved aging by the end of the century. Since Larry expects to be explore the universe on Von Neumann probes, he also implicitly expects aging to be a solved thing in 2100. The interesting fact here is that he postulated this in 2015, when the idea of solving aging and the related death was not so prevalant across the world. He was way ahead of everything. Wondering what his newest views are around this.
I think solving aging would be the biggest achievement of humanity. I am sure superintelligence will help with this. I can't wait for it.
Microsoft's Mustafa Suleyman reveals Larry Page told him at SpaceX that by the end of the century he’d be exploring the galaxies at the head of a von Neumann probe
“Yeah, I remember in 2015 being in the canteen at SpaceX.”
“And we had our first meeting of the Ethics and Safety Board. So when we sold DeepMind to Google, we made it a condition of our acquisition that we would have an Ethics and Safety Board with independent representation on the board."
"One of those independent members, because Google had to agree, which was obviously Larry, Sergey, and David Drummond at the time. And then me, Demis, and Shane had to jointly agree who those three independents would be. And so we jointly picked Elon as one of them.”
“No idea if that had an influence on the founding of OpenAI or not. But so then we had our morning session and we were sitting down for lunch and I was sitting next to Larry. You know, obviously he was suspicious of me as well because he was like, you know, you’re one of those speciesists.”
“And he said, you know, I think that what you don’t understand, Mustafa, is that by the end of the century, I’m going to be exploring the galaxies at the head of a von Neumann probe, which means a von Neumann probe is a kind of self-propelling kind of robot that is able to gather energy from space as it’s careening through the light years.”
3
238




















