Irvin Sha
297 posts
Irvin Sha
@irvincsha
I deleted TikTok so now this is how I rot my brain | Corp Dev + Investor Relations at AWX | ex-HOOD, ex-JPM FIG IBD | Mostly harmless
Hillsborough, CA Katılım Mart 2008
238 Takip Edilen122 Takipçiler
Fraudsters and spammers are so prolific, I can't help but wonder how well they'd do if they went legit. So much wasted time on both sides.
English
@pitdesi @theallinpod I was actually pretty shocked when I heard this. I didn’t think he had the mental flexibility to know how to pander to this particular audience.
English
Trump on @theallinpod:
“What I want to do and what I will do is you graduate from a college, I think you should get automatically, as part of your diploma, a green card to be able to stay in this country.“
He’s doing the exact opposite, brain draining the US.
The Wall Street Journal@WSJ
The Trump administration will limit how long a foreign student can remain in the U.S. on a student visa under a new regulation announced on Thursday, increasing the hurdles for foreign students and universities’ ability to recruit them. Read more: wsj.visitlink.me/fpk5vv
English
@TheStalwart And the norms of their job require them to regularly write a bunch of stuff that they don’t actually give a F about
English
A thing to remember about AI-generated LinkedIn slop is many of those people (marketing people, corporate division chiefs, comms and PR professionals) are objectively among the most educated and literate people to have ever walked the face of the earth.
English
LOVE THE NO
All founders should learn to Love The No.
— The No brings clarity. You know where you stand; things can only get better from here.
— The No sharpens the pitch. Every rejection tells you what didn’t land.
— The No filters fast. A quick No beats a slow “Maybe” every time.
— The No builds resilience. The founders who win are the ones still standing after 99 of them.
— The No isn’t the end. It’s just not-yet, or not-them.
To get to Yes, you’ve got to Love the No.
English
@tonyzzhao This is sick! Now add fitted sheets to the set of tests.
English
Great thoughts @MBGilroy . Would just note that the “sell-side” leak hypothesis is fair, but bankers have been knocking on doors about this opportunity for months. So it’s less about forcing people to pay attention, more about forcing people to hurry up and put in a bid.
That said, the credible buyer set is really small.
English
Stripe / PayPal...
Deal: Stripe + Advent buying PayPal for $53.4B, or 28% premium to yday closing price. The two buyers would own equal stakes in the business and "PayPal hasn’t responded to the offer" which we keep reading...
English
this is downstream of the fact that the number of reporters who report being republicans has plummeted since the 80s
Reporters who actually do reporting in the form of words -- as opposed to talking heads on FOX -- are overwhelmingly left-leaning.
The huge win for repulicans would be to STOP doing talking head nonsense, and inveset in actual reporters like NYT and WashPost have
English
An interesting read. I agree with a lot of what he’s saying. The fundamental issue is most people really suck at discerning who they should trust/listen to.
The criteria are generally unobjectionable, not sure I would have chosen the same set of people that he did. In practice, my hunch is that most of society will just blindly trust people with big followings and no editorial oversight or obligation to speak the truth.
Marc Andreessen 🇺🇸@pmarca
An epoch-defining essay. Wholeheartedly agree. aaronrenn.com/p/the-new-trus… @aaron_renn
English
Good post from @Alfred_Lin. I view interviews as a really suboptimal way of hiring, in search of a better alternative. It’s like deciding to marry someone based on a single date, having your friends talk to them, and maybe seeing what their exes say.
I ask my candidates a similar question to Alfred’s namely, “What is something that you are horribly, embarrassingly bad at? And why?” Critical observers will note that this is a modest tweak to just asking about a person’s weaknesses. The key difference is that I’m pretty explicitly telling them not to give me a strength dressed as a weakness. If it’s a weakness that nets positive, then there’s a real question as to why you’d consider that to be embarrassing.
More importantly, I’m testing to see how quickly they can show vulnerability, and get to transparency. Especially when you work with a small team, this is really important.
Alfred Lin@Alfred_Lin
English
Michelin star restaurants are overrated
LA has better food
Sheel Mohnot@pitdesi
TIL the Bay Area has 44% (7/16) of the Michelin 3 starred restaurants in the US New York has 5, LA has 2
English
This is good. Given all of the alarm bells raised re: AI safety, particularly by members of the major labs, I hope that it actually goes somewhere beyond generating clicks.
“This is a pivotal moment in human history. Artificial General Intelligence (AGI), a system that exhibits all the cognitive capabilities the brain has, is probably only a few short years away.”
This phrase in particular caught my attention. I’ve really struggled with the way that people discuss the “intelligence” of AI, as the basis of measurement generally seems to be whether it thinks like a human. That does not strike me as a good yardstick.
First, intelligence has many dimensions beyond “smart vs dumb,” which seems to be the way that everyone on this platform talks about it. You can have a lot of intellectual horsepower, but a limited set of knowledge to apply it to. You can be really good at organizing disparate lines of thinking, coming up with novel solutions, learning new skills, being self-reflective, etc.
AI is already better than most humans at many things. The amount of knowledge it has at its disposal already dwarfs that of almost anyone on the planet. Does it have to be at/above our level on every dimension before we think AGI/ASI is achieved?
Secondly, it’s not obvious that all human cognitive capabilities are something that we would desire in AI. If you consider emotion to be a cognitive feature of ours, would we really want that in AI? What about instincts, eg self-preservation?
Tl;dr - I’m nitpicking on the goalposts for believing that intelligence has been truly achieved. My view is that grounding the definition in how humans think is not the way to go.
Demis Hassabis@demishassabis
English
English
About two weeks in and Robinhood Chain’s growth continues with companies like @alchemy @morpho @rialto_xyz and @trmlabs.
👉Over 52.5M transactions
👉Nearly 1M total addresses
👉Over $323M protocol TVL
👉$3B+ in weekly DEX volume (per Bernstein)
Keep building.
English
@benjitaylor Any reduction in the number of bot posts telling me how to use Claude like Karpathy is a blessing
English
Extremely positive response to the new 𝕏 algo. Great to see!
English
@linasbeliunas (Mostly) yep. And the big question is how all the neobanks will perform once they’re regularly on the receiving end of the Fed and OCC’s proctology exams.
English
And then all neobanks end up being just banks.
Luka Ivicevic@lukaivicev
Building a neobank sounds glorifying until you realize its all KYC and compliance.
English