Amit Patel
@bloopmit
Tokenmaxxer trying to figure out the future of work and markets
You might like
Seeing institutions drag their feet on AI is very interesting and frankly a little exciting. Ironically the guys claiming to be long term analysts are the ones most dismissive of making investments in AI now. The “alpha window” on this may be a lot more meaningful than the one…
@blueprintsmb22 captured a lot of the nuance of AI in investment research very well. It's easy for software coders to write long think pieces about how AI is going to destroy Wall Street jobs. A software developer hunched over code conducting verifiable work (either the app…
Talk about getting framemogged
this guy’s entire bit is trying to make food with his cow in the kitchen. absolutely tremendous content
Building environments that require out of distribution thinking requires infrastructure that can support long trajectories with intermediate validation steps. Looking forward to publishing more on the environment and research end soon!
RL envs are only useful if training on them improves a model - but those experiments are harder the more difficult (and useful) the environment is. @bloopmit and I have been working on some fun benchmarks and envs, and training many models over them. We're opensourcing our…
Looks like my employer is going to be billing Claude a lot
A buddy of mine ran a large part of the DoorDash engineering org. When an engineer caused an outage, they knew exactly what the revenue cost was, and that revenue cost was hung on that engineer. They didn't do fines, per se, but it went into performance reviews, and it…
I have no idea what Rentech does but I am willing to bet it’s not “discovered LLMs in the 90s”. That or they found the most compute efficient machine intelligence anyone has ever discovered 30 years ago
the conspiracy theory I most strongly believe is that rentech discovered LLMs in the 90s and used them to dominate markets for 20 years+ before everyone else stumbled onto them for language generation Peter Brown and Ilya both had Geoff Hinton as their PhD advisor.
If the Chinese Communist Party really hired McKinsey to destroy us we have legitimately never been safer
Housing is a centuries-old technology and should be basically free by now, just like food/water/electricity. It's pure insanity that we've made it a scarce unproductive fake "investment" asset and allowed it to take up an insane amount of people's budgets
They didn't
The Mexican government has released pictures of what happens to a human after 2 years of playing a popular game “league of legends”. It's not looking good.
Whartonite seeks code monkey —> Code monkey automates Whartonite
GPT-4 competes against human MBA students in generating innovative ideas. Results in a resounding victory in favor of GPT-4 within the top 10% of ideas. Final score: 35 - 5. Source: WSJ.
“gang gang yes yes yes gang gang mmm ice cream so good gang gang”
The last thing Paradigm backed before removing all mentions of web3 from their website
“The firm appears to have removed crypto/web3 mentions from the front page of its website earlier this month, suggesting a rebrand that comes after high-profile collapses in the sector and rising regulatory uncertainty in the U.S.”
Should have shown this to my parents when I was in school
WSB can finally play the fed directly in rates
ROBINHOOD EXPECT TO LAUNCH FUTURES TRADING BY END OF 2023 $HOOD
ROBINHOOD EXPECT TO LAUNCH FUTURES TRADING BY END OF 2023 $HOOD
United States Trends
- 1. Happy Friday the 13th N/A
- 2. #RISERCONCERTD1 N/A
- 3. Crew-12 N/A
- 4. Brent N/A
- 5. International Space Station N/A
- 6. JASPER KINDLED FIRSTRISE N/A
- 7. Human CAPTCHA N/A
- 8. Chibi N/A
- 9. Perle Labs N/A
- 10. Falcon 9 N/A
- 11. #HudDay N/A
- 12. hudson N/A
- 13. Zimbabwe N/A
- 14. General Yao Guang N/A
- 15. $HTC N/A
- 16. #TheTraitorsUS N/A
- 17. #thepitt N/A
- 18. #criticalrolespoilers N/A
- 19. Jill Scott N/A
- 20. Tubi N/A
Something went wrong.
Something went wrong.