Founder at Stealth | AI @Stanford | a16z Speedrun Founder | EB1A Recipient
Palo Alto, CA
Joined July 2015
- Tweets925
- Following965
- Followers372
- Likes1.5K
Srijon Sheikh retweeted
Did office hours yesterday with a very smart but nerdy founder. When I suggested he might want a more outgoing cofounder to do sales, he said he didn't, because most of his sales would be to agents. I had to admit he could be right.
Srijon Sheikh retweeted
I had no idea EB-1A approval rates had dropped this sharply.
EB-1A is the US green card for people with extraordinary ability (founders, scientists, artists, etc.).
Approval rates fell from ~75% to ~32% since early 2025, while O-category visa approvals stayed around 90%.
Srijon Sheikh retweeted
I sincerely hope that a version of what @paulg wrote here is true, that kind people succeed as I have gravitated towards kind colleagues & now founders, all my life
Srijon Sheikh retweeted
Autonomous consumer agents (Grokbot, Instinct, Town, Muse, etc) are the biggest win for consumer pricing power
Airlines, hotels and retailers are 100% going to need to revise their policies and strategies when a material % of their user base is:
- monitoring for price changes to bookings post purchase
- capitalizing on “lowest price guarantee” claims
- automagically requesting late arrival airfare fees and other service blusters
- notifying a consumer to hold off on making a purchase because a retailer historically launches a sale the following week in a season
- negotiating a quote received from one home services vendor across a sea of alternatives
Agents begin the march towards consumer abundance
Srijon Sheikh retweeted
Do you have an AI confession?
This year, I collected 400+ handwritten secrets from strangers about their relationship with AI.
Check out the archive here.
WWW.WHATWETELLAI.COM
Designed by my brilliant friend, designer, and collaborator @arin_pantja
Srijon Sheikh retweeted
Toast might be one of the few public SaaS companies that have launched a true "AI-native services" offering.
Toast IQ Grow is a $499/month AI marketing service that pairs an AI agent with human support.
It is the fastest-growing new offering Toast has ever launched
Srijon Sheikh retweeted
A lesson I just taught a Flexport employee is that there are two lenses you can look at the world through:
1) Doing more work is bad
2) Solving problems is how you make money
Always and forever use lens 2. More work is good if it lets you solve important problems.
Srijon Sheikh retweeted
An initial startup idea can't usually be both grand and precise. In practice they're usually either grand and vague or precise and small. Precise and small is better. You know who your initial users are, and you expand outward. With grand and vague you can't even get started.
Srijon Sheikh retweeted
As I get older, I’ve become convinced that training yourself to be unemotional is the number one skillset for success in business and life. Getting mad, letting your pride interfere, being jealous, wanting to score points backfires 100% of the time. Exercise and time is a key.
Srijon Sheikh retweeted
After seeing Kunal Shah leave CRED for WhatsApp, I had a few realizations from my entrepreneurial journey.
It's very important to know why you are starting a company, and a company is just a part of your "why" as a whole for your life. Starting a company is a tool for you to achieve your why.
Reason 1: Starting a company to make money. Here, your "why" is to make money. If you can make more money in a job, go for it. But if you see a good business you can build, go build it. Your filter for that decision should be, "Where can I make the most amount of money for myself?" Also, remember, a company's revenue is not your money.
Reason 2: Starting a company to solve a problem. Here, your "why" is to solve the problem, not to make a lot of money. You started a company to solve the problem that you wanted to solve. I will give an example here of @demishassabis. He sold DeepMind to Google, and his explanation was that he started DeepMind to solve the AGI problem. In the middle of it, he found himself lost in fundraising and managing the company. It made him go further and further away from focusing on solving the problem, so he sold DeepMind to Google so Google could take care of the finances and the management, and he could go and work on solving the problem. If you are a founder who started a company for this reason, it is okay for you to take up a job in an existing company that is already working to solve the problem. Starting a company is not always the right move to make here.
Reason 3: Starting a company for status. This is probably the worst reason to start a company. Don't do it. Get your status from other things in life. Even if you start it, you will be focusing more on becoming the "IT"/"Famous" company than focusing on the business side. Most founders who do this fail the most.
People don't recognize this, but it's better to do a job than to stay stuck in a bad business. Business is hard, and if the pain is not worth it, why are you spending time on it? Start something new and better with what you know now, or get a job to learn and earn faster.
I have seen my own companies that I started, and the companies started by people I know. Many failed after 2 years because starting a company is more about things you don't control. When you are new to this, you ignore them and think that if you work hard enough, you will become successful.
What you work on is very important. You are grinding to build your $10k agency day and night, while your friend built a tool that helps solve a profitable problem for other businesses and made a recurring business from it with one-third of the work you do, and he makes more than you.
Srijon Sheikh retweeted
Every piece of advice is a reaction to something that came before.
"Validate your idea before you go all-in" was a reaction to MBAs raising millions of dollars based on nothing but a business plan.
Today, young builders take that advice and quickly "invalidate" every idea they come up with.
Now the missing ingredient is conviction 🤷♂️
Srijon Sheikh retweeted
Let me explain exactly why Amazon would rather hand a finished, well-tested film to a rival studio than release it themselves.
Artificial is done. Guadagnino directed it, Garfield plays Altman, and the L.A. screenings got a warm reception. This is the same director whose last movie, Challengers, did about $90M worldwide. On pure commercial logic you release this: an awards-season prestige drama landing while AI is the only story anyone's paying attention to. It prints money.
Now look at the other side of the ledger.
In February, Amazon committed $50B to OpenAI: $15B up front, $35B more if conditions are met, with AWS becoming the exclusive third-party cloud for OpenAI's frontier models. That last clause recurs for years and runs into the billions annually. Bezos and Altman are close enough that Altman showed up to his wedding.
So run the trade. Best case, releasing the film clears Amazon a few tens of millions after theatrical splits and marketing. The downside it creates: an unsympathetic portrait of your $50B partner playing in theaters worldwide right as that relationship scales.
Here's the tell. Amazon greenlit this knowing exactly what it was. They saw every early iteration of the script before Guadagnino even boarded. Then February happened. The same asset that was a green light in 2024 turned into a liability the moment Altman became the counterparty on a $50B deal and AWS's biggest new customer.
Walking away from a $40M film to protect a relationship worth a thousand times that is the easiest call in the building. The trade was never even close.
BREAKING: Amazon is canceling the release of Sam Altman's biopic "Artificial"
The film, starring Andrew Garfield, completed production and underwent positive test screenings, but Amazon MGM Studios dropped distribution rights amid OpenAI and Amazon's deepening $50 billion corporate partnership
The producers are currently shopping the completed film to other studios for distribution
Srijon Sheikh retweeted
a YC founder i'm working with was worried because one customer had already built a simple version of his product in-house.
i told him that's not a bad sign, and it's actually the best proof point he has.
if a company is spending expensive engineering time on the problem you're solving, that validates a bunch of things at once:
- the pain is very real
- other vendors are not good enough
- many companies likely have the same pain too
