Pranav Sriram (@pranavsriram1) 's Twitter Profile
Pranav Sriram

@pranavsriram1

Interested in minds, machines, art, and math. Founding Eng @ Traversal, @Stanford alum, author of Olympiad Combinatorics

ID: 1386101913679134722

calendar_today24-04-2021 23:42:32

1,1K Tweet

423 Followers

485 Following

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Cynical Publius The stronger your argument, the less afraid you are of people hearing the other side. Censorship is an admission that your position cannot withstand scrutiny.

Armand Domalewski (@armanddoma) 's Twitter Profile Photo

The most important thing San Francisco politicians can do to resist Trump is get our own house in order. San Francisco should be a promise of what America can aspire to be, not a cautionary tale for what it must avoid becoming

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

𝖦𝗋𝗂𝗆𝖾𝗌 ⏳ @suicideforclout Rare to see <detailed argument>, <specific counterexamples>, <immediate agreement from OP> on this app. Nice to see when dunking is the default!

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Re the ongoing skilled immigration debate - several things can be true simultaneously: 1. American leadership in science and innovation vitally depends on our ability to attract the best and brightest minds from around the world 2. There are legitimate issues with the H1B

Pavel Durov (@durov) 's Twitter Profile Photo

Humans are hardwired to copy others by evolution — for a good reason. Mimicking those around us has been the most efficient survival strategy for millions of years. However, the world is different today. We now live in an age of rapid change — what used to be efficient becomes a

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Bryan Cheong I used this as an example in a lecture about the notion of “state” in AI algorithms (particularly search and RL). Fully specifying state in chess is a tricky exercise, due to en passant, castling rules, 3-fold repetition, repeated checks, and the 50-move rule

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Interestingly, Claude, GPT 4o, Grok, and Gemini all get this wrong Of the models I tried, only o1 gets it right (18s of thinking)

christoph (@halalcoholism) 's Twitter Profile Photo

Let’s be clear, most of us who’ve been saying “cancel culture is bad” meant let’s not try destroy people over misunderstandings & minor transgressions. It never meant that it’s fine for someone who was posting NeoNazi rhetoric just a month ago to be working in the government.

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Nice thread exploring some interesting implications of finite precision and computability theory A lot of seemingly simple things in computer science are actually really, really hard in a fundamental sense

Pranav Sriram (@pranavsriram1) 's Twitter Profile Photo

Interesting issue in 4o - whenever starting a markdown bold with **, the subsequent word has the first two letters repeated (e.g. PrPremier, CrCritical, etc.) Anyone else seeing this recently?

Interesting issue in 4o - whenever starting a markdown bold with **, the subsequent word has the first two letters repeated (e.g. PrPremier, CrCritical, etc.) 

Anyone else seeing this recently?