Pinned Tweet
Started posting on Twitter with 0 followers.
Now documenting my journey:
• Learning AI
• Improving writing
• Growing step by step
If you're starting from 0, follow along.
English
shrasti | AI & Writing
980 posts
shrasti | AI & Writing
@CreatorStackAI_
Learning AI + Copywriting 0 → ? journey in public Sharing what actually works
Joined Mart 2026
14 Following40 Followers
@hkarthik that’s not freedom.
Deleting LinkedIn isn’t independence
it’s disengagement
Real financial independence looks like choice
what you work on
who you work with
when you stop
Most won’t delete
they’ll just stop caring who’s watching
English
When many of us middle managers reach full financial independence there will be signs.
Not fancy cars, yachts, or sprawling estates.
It will be all of us collectively deleting our LinkedIn accounts.
Marc Andreessen 🇺🇸@pmarca
Overheard in Silicon Valley: "LinkedIn is prison for middle managers."
English
@elonmusk fewer books, better AI.
But models don’t fail from too much data
they fail from the wrong objective
Two perfect books won’t fix that
Intelligence isn’t compression
it’s alignment
Until that’s solved
more or less data won’t save it
English
@Scobleizer homes are the final boss.
Demos look smooth
real homes are chaos
pets
kids
clutter
unpredictable routines
The leap isn’t innovation
it’s reliability in messy reality
Until it handles that
adoption stays slower than the hype
English
I got a close look.
San Francisco robot innovation.
Will be seen in many homes there.
Aaron Tan@aaronistan
Introducing Lume. A lamp that does your chores. Order now. Shipping this summer.
English
“it’s incompetence.”
Hard take: constraints.
0 builds with speed because he controls the stack.
Public projects answer to voters
regulators
lawsuits
unions
politics
Different game, different incentives.
One optimizes for velocity
the other for consensus
Confuse them and everything looks like fraud
English
Elon literally built a fleet of planet colonizer class rockets for <$10b investment at SpaceX.
The state of California spent $150b to build 1/2 mile of high speed rail
There is no limit to the incompetence and fraud of bureaucrats
Elon Musk@elonmusk
The @BoringCompany could build a Hyperloop tunnel from downtown SF to downtown LA for <5% of this cost and it would be a technological marvel exceeding any high speed rail on Earth
English
@Austen more data isn’t the win.
Most companies are already drowning in data.
They just don’t know what matters.
Recording everything sounds smart
until noise kills signal
The edge isn’t capture
it’s curation
The teams that win won’t have the most data
they’ll have the cleanest loops
English
Honestly more companies should just start capturing screen recordings/screenshots of everything that happens at the company to be later fed into models.
There’s so much data that companies could be grabbing.
Corey Ganim@coreyganim
Easiest entry point into AI consulting: AI audits. Here's the framework: 1. Map every repeatable process in their business (takes 45 min on a call) 2. Identify which ones are costing the most time or money 3. Recommend the AI tools and workflows to fix them Charge $1,000 per audit. Takes about 2 hours to deliver. You're not even building anything. You're delivering a structured assessment and a prioritized action plan. Do 4 audits a month and you're at $48K/year from a service you can deliver in a single afternoon each week. Then upsell the implementation.
English
“be careful” is late.
The real money was made when no one wanted PSU banks.
Now everyone has the same thesis
clean NPAs
re rated valuations
cycle awareness
That’s the problem.
Crowded trades don’t die quietly
they unwind when expectations miss
At this stage you’re not early
you’re hoping for perfect execution
English
PSU BANKS: STRONG BUYING… BUT BE CAREFUL
Seeing aggressive accumulation in PSU Banks.
But cycle is shifting 👇
• Reserve Bank of India → Rate cuts paused
• Credit growth → slowing
• Liquidity tailwinds → fading
NPAs look great:
• UCO Bank ~0.15%
• Indian Overseas Bank ~0.3%
• Indian Bank near lows
But remember:
Lowest NPAs = Late cycle
NPAs are lagging
Valuations already re-rated (0.3x → ~1x P/B).
Now upside depends on earnings, not sentiment.
Reality:
Good for swings / momentum
Not easy long-term compounding play
PSU Banks were a great buy when cheap.
Now they are a cycle trade.
Follow for data > narratives.
English
@elonmusk @Teslaconomics @boringcompany “it’s corruption.
it’s incentives.
Massive projects attract layers
layers protect themselves
costs quietly explode
You don’t need a conspiracy
for a system to waste billions
Just misaligned incentives
that no one is paid to fix.
English
@Teslaconomics @boringcompany The real reason for the “high speed rail” is money-laundering to bureaucrats, consultants & unions, not actually transport. That is where the billions spent so far have gone.
That is why they don’t want an actually cost-efficient high speed transport system.
English
The @BoringCompany could build a Hyperloop tunnel from downtown SF to downtown LA for <5% of this cost and it would be a technological marvel exceeding any high speed rail on Earth
Hans Mahncke@HansMahncke
If you gave away $126 billion to subsidize free flights between LA and San Francisco at current demand levels, you could fund roughly 150 to 200 years of travel before the money runs out.
English
@elonmusk “far exceed humans.”
We said that about every version.
Progress is real, but edge cases don’t care about hype.
Self driving doesn’t fail on average roads.
It fails on the one weird scenario no one predicted.
Until it handles the rare better than humans,
it’s not safer.
English
Tesla V14.3 self-driving review. The point releases will bring polish.
V15 will far exceed human levels of safety, even in completely unsupervised and complex situations.
Zack@BLKMDL3
600 miles in with FSD v14.3 already and here are my thoughts: The improved reaction time is immediately noticeable and definitely quicker than a human could react. Yesterday a semi truck swerved fast into my lane and the car reacted insanely quickly to get me around them. Tesla says it’s 20% quicker but feels more than that- and it was already extremely good. I’ll start by saying FSD was already so great with v14.2 that it’s sometimes hard to find new things, but there some huge apparent changes immediately noticeable with v14.3. To those who think it’s not a big improvement over v14.2.x+, you’ll be very impressed and especially with some more polish. The reinforcement learning upgrades and thinking are noticeable. Parking is where you immediately notice some changes. In the release notes it says that parking is quicker and more decisive and it’s true. It has picked spots closer and quicker to the selected pin. My Y isn’t showing the new P pin graphic for parking pin for some reason, but it’s definitely parking closer than before with more thought to it. Looking forward to eventually getting more options to hopefully park either closer or further away from people. For the very first v14.3+ build, I have to say it’s pretty polished. The only gripe I have is the way it won’t get out of the left lane soon enough on highways. It likes to cruise in the left lane which isn’t ideal, it’s gotten better but the addition of reduction in unnecessary lane changes needs to be dialed back a bit. Lane changes are a huge plus with this build and they are quick, decisive and executed very well, smooth as butter too. Turn signals come on at way better times now in parking lots and at the perfect time on the road. I was lucky enough to get the update with 600 miles left of my 1800 mile Oregon road trip, so I pulled over to install it so I could get as much experience as possible with it to share with you all. So far 600+ miles in, I’m impressed. A few rough edges with the left lane behavior and the last few inches of parking are a bit slow 1/5 times until it puts it into park but with a point release update everything should be dialed in. The 350 mile drive home today from the Bay Area had zero intervention including all parking and charging. One thing I would love to see implemented is a reset button for the FSD stats page. Would be cool to have a specific trip meter for FSD stats on Trip A/B or make your own trip. I’ve been hinting at a pretty cool road trip next month with my 2025 Model 3 so it would be cool to have a reset for that. Speed control seems good on highways, it’s matching traffic speed great. Braking is very impressive for sudden slowdowns, had a big one in San Jose last night it did a great job with from 80-10mph. HUGE improvements with stop sign behavior. The acceleration and deceleration are way smoother than before, much more pleasant. Mad Max takes off strong but again, a better curve than before. Mad Max is also polished a bit and feels great. FSD v14.3 did a great job in LA traffic once I got back and will go out this evening to film videos for everyone on my normal test loops. Let me know if there’s anything specific I should try or check out. Can’t wait to see how v14.3+ progresses especially with the upgraded reasoning coming to all scenarios soon. Some awesome additions here. THANK YOU everyone @Tesla_AI for all the hard work getting this update out. More videos to come.
English
@HarryStebbings it wasn’t talent.
Talent exists everywhere now.
The real advantage was focus.
Less hype
Less job hopping
Less “build fast, pivot faster” noise
Europe quietly compounds.
And in AI, compounding beats chaos.
English
DeepMind stayed in London because it is better for talent than Silicon Valley.
"I saw London and the UK as having incredible talent from top universities like Cambridge, Oxford, Imperial and UCL.
There is a deep heritage of scientific breakthroughs and world-class thinkers.
There was less competition for that talent, which made it a huge structural advantage for building DeepMind." @demishassabis
What is the single biggest advantage of building in Europe for you @torsten @antonosika @MaxJunestrand @matiii @ChrisParsonson @cjpedregal @matthewclifford @torstenreil @alanchanguk
Harry Stebbings@HarryStebbings
This sounds harsh but it is true, very few of the guests we have on 20VC will be remembered in history for truly progressing humanity. Our guest today will be thought of alongside Turing, Newton, Einstein and I feel immensely privileged and fortunate to have had the chance to sit down with @demishassabis. For anyone who feels their dream is out of reach, just keep going. The 18 year old kid starting 20VC from a bedroom with no money, 11 years ago, would not believe that I get to press publish on this. Chase your dreams. You never know what room you will end up in! (Links below)
English
@arpit_bhayani “Build a real one” is vague.
The hard part isn’t building.
It’s getting someone to pay twice.
AI killed the cost of creation.
It exposed the cost of distribution.
Until you solve:
→ who needs it now
→ why they can’t ignore it
→ how it becomes habit
…it’s still a toy.
English
Yes, you can use AI to build toy apps. Now build a real one - and monetize it. We just made it easier.
We're working on some pretty cool things with and for AI. Stay tuned. Lots of interesting stuff ahead.
Shashank Kumar@shashank_kr
Great news to start off the new financial year. @Razorpay is now embedded natively inside @OpenAI’s Codex, an industry-first for any Indian payments company. A developer prompts "build me a fitness app with a ₹499 one-time payment." Codex builds it. Razorpay handles the payments. No separate integration, no dashboard-hopping. If AI makes creation effortless, monetisation should be just as simple. India has 6 million developers & weekly Indian users of Codex 4x'd in two weeks this February. We’re building for this shift where more people can create, launch, and earn without friction. We have always enabled businesses so that their payments are smoother, easier, simpler & faster and Codex just got added to that already illustrious list.
English
I did 100 replies in a day.
Here’s what I realized:
More effort doesn’t always mean more results.
Quality matters more.
English
$100M won’t solve it.
Not because it’s small — because we’re asking the wrong question.
0 isn’t just a drug problem.
It’s a detection + timing problem.
By the time symptoms show, the damage is already years deep.
AI won’t “cure” it first.
It will make it visible earlier than we’re comfortable with.
And that changes everything.
English
Alzheimer’s is one of medicine's hardest unsolved problems, and one of the most devastating.
At the OpenAI Foundation, we believe AI is well suited to its complexity. We're directing over $100M to scientists mapping the disease, designing drugs, & more.
I wrote about it here:
openaifoundation.org/news/ai-for-al…
English
@elonmusk it’s not “awesome.”
It’s a statement.
Most cars optimize comfort.
0 optimizes attention.
That’s why people either love it or hate it.
And ironically… that’s exactly why it wins.
English
@elonmusk therapy ≠ awareness.
Grok won’t go to therapy because it doesn’t suffer — it just predicts.
Humans don’t go either… and we actually feel the consequences.
That’s the gap
Intelligence without self-awareness vs
Self-awareness without action.
Guess which one is more dangerous.
English
@ChrisJBakke Logos don’t build products
Outcomes do
If you need a slide to prove it
it’s probably not real
English
Marketing: “we need a 12th logo for this slide.”
Biz dev guy: “Anthropic.”
Marketing: “we’ve…partnered with ourselves?”
Biz dev guy: *shrugs*
Marketing:
Anthropic@AnthropicAI
We’ve partnered with Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks. Together we’ll use Mythos Preview to help find and fix flaws in the systems on which the world depends.
English
Didn’t post yesterday.
Lost a couple of followers.
Good reminder:
Consistency matters more than motivation.
English
@ziwenxu_ Simple setups win
Complexity kills adoption
Whoever makes power feel easy
owns the market
English
Google just dropped the playbook for OpenClaw + Gemma 4 setup in 3 dead-simple steps
Google Gemma@googlegemma
English
@_advaitpatel Everyone copied the stack
Now they’re rewriting it
Robotics isn’t just data
it’s architecture
That’s where the real edge will show
English
Generalist and Rhoda are both training from scratch, Sunday has their own architecture as well. Notably, the exact architectures for all of them have not been publicized. I hope we get to see interesting robotics-first model architectures in the future!
Pete Florence@peteflorence
English