The Ghosts We Inherit: India's 'New Money' Story
My mother reuses tea leaves until the water runs clear. My friend's father keeps a wooden box in his closet, with carefully folded bills arranged by denomination. Another's maintains a small diary tracking every household expense down to the last rupee. These aren't quirks – they're battle scars from a generation that survived on less.
We are their children. We carry their financial trauma in our wallets, even as we tap them against sleek payment terminals at craft coffee shops.
big!!
Looks like Jev will make it 500x easier + cheaper to bake human judgement and guardrails into harnesses.
I’m expecting reliable intelligent workflows to now truly flourish.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
modern parenting in a metro can feel like a very well furnished prison.
I know how that sounds. 3/4BHK (and EMI), cook, car, two incomes. Nobody is coming to feel sorry for anyone. But it still feels heavy.
my parents raised me where almost all neighbouring families knew my name. there was always some neighbour who cared to feed me lunch when mom was late. summers in grandparents' place. nobody was paid for any of it. that was the village. it was free and it was everywhere.
then folks (who are now in their late 30s/40s) moved to Bengaluru/Gurugram/Pune for work, kept the kids, and lost the village.
so now two adults run an operation that a dozen used to share. and the way you run it is : you buy the village back.
nanny, daycare, a school that picks up at 7 30 am and drops the kid back by 3-4pm, the extra bedroom for the grandparent/nanny.
get the drift? a kid in a metro costs more precisely because every hour of care that used to be free is now a line item.
the part I keep chewing on is the double income. it is not greed. it is the only way the math even works. but it comes with its own bill - two people on 9pm calls, family time under attack, and a mental load that no amount of hired help absorbs.
the alternative is one parent pauses. that pause usually lands on the mom. not because anyone sat down and chose it, but because the salary gap, the maternity leave, and the fifteen "who will pick up" moments a week all point the same direction. I don't think that is a choice families are making. it is a default the city hands them.
and the third option - grandparents move in - means two 68 year olds give up their friends, their evening walks, their own city, to be free labour in one they didn't pick.
every version costs someone. money, a career, presence, or a parent's retirement. there is no configuration where nobody pays. we just get to pick who.
Families raising kids in the cities are today one of the most under-served segments of society - they have all the comforts without the care young families need.
big!!
Looks like Jev will make it 500x easier + cheaper to bake human judgement and guardrails into harnesses.
I’m expecting reliable intelligent workflows to now truly flourish.
@championswimmer@Railway I think we’re now on Vercel purely for developer experience on staging builds + CDN / image optimisation for our frontend.
Migrated out everything else onto Cloudflare. I use Railway for ancillary stuff.
Will take you up on this when we’re cooling down a bit!
Love both products, but I’m starting to think @Muse is materially better than Instinct. Like how it pops up a mini browser when it hits something it can’t do. Gives the user back control instead of stalling. Thoughtful, human-centric AI design.
fantastic product (!!)- you're hitting on something bigger than Instinct: it's not just the giants who can eat your lunch anymore. anyone can.
three years ago, building something was grabbing a couple of people, raise a little seed, put your head down, and ship in three or four months. That window was the moat. Three or four months of creativity with a little wind at my back, enough to build momentum so that by the time anyone cloned us, we were already two steps ahead.
Now, two things scare the shit out of me.
first, (online)social connectivity is at an all-time high. ideas propagate wider and faster than they ever have, and they reach the right people.
second, AI has 10-20x'd output, and it did it without seed capital. seed money was its own barrier. you had to know high-net-worth people, pitch them, and close the round - now I can point AI agents at any app, catalog every feature, and have a working clone in a day or two max. and we're at the very beginning. what takes AI days today will take hours in a few months.
so where's the moat? brand. trust and loyalty, and most likely to individuals rather than companies. If @tferriss launches a productivity app, you take it seriously in a way you never would with a random App Store listing. If @hubermanlab ships a longevity tracker, you're checking it out. In the short term, brand trumps everything.
which brings it back to Instinct. there's no question every major AI player has already spun up teams to build a competitor e.g. Muse. that's what makes this moment so exciting: the barrier to entry is next to nothing. It's also what makes it terrifying: the barrier to entry is next to nothing.