carlbeek (@CarlBeek) header
carlbeek (@CarlBeek) profile picture

Systems researcher & engineer · Previously research @ethereum.

Joined December 2011
Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)Photo by carlbeek (@CarlBeek)
First they came for free users, and I didn’t speak up as I was on a paid plan Then they came for the power users, and I didn’t speak up as I had free resets Then they nerfed the $200 plan, and I didn’t speak up as I was grandfathered in Finally they came for me, and there were no tokens left to speak up for me
Heads-up: stacking paid accounts on one machine for extra Codex usage can now get you locked out. OpenAI's new help page on card and device limits: - One device can only use so many paid accounts (Go, Plus, Pro 100/200/500) - Hit it and you see "Device limit reached" at sign-in, or get signed out later, even if the purchase went through - Signing out of another account doesn't free a slot - Business, Enterprise, Edu and Free don't count toward it This lands the same week multi-account switching shipped on mobile, so check before you buy a second sub. The sanctioned route to more usage is credits under Usage & Billing in the Codex app. help.openai.com/en/articles/…
2
383
I'm contributor #194 to the MOSS Social zkLogin ceremony. d0a7975e7e413fe6372b5e0ea8d19028c2fe60b8f64d98a0cfc830e1ffcfe4d4 The more people who join, the stronger it gets. Contribute your entropy here: ceremony.megaeth.com
1
3
352
1/ I built a bot to get rich without ending civilisation in @danrobinson & @justinwangx’s Pace. Unfortunately, the other player also gets a vote on the civilization bit. xitter.cf/justinwangx/status/210…
Can you pace the frontier? @danrobinson and I made a multiplayer game inspired by a game theory paper from @andrewjkoh where you play an AI lab. Racing ahead increases your profits… but also the risk Try to beat the computer — or your mutuals online :) paradigm.xyz/research/pace/
2
1
9
688
3/ Some lucky bastards played one great game and retired. I kept playing because I wanted statistically robust results. The optimal stopping strategy was apparently closing the tab before anyone found out you were lucky.
1
3
122
4/ Anyway, I sincerely hope Sam and Dario are bots with several thousand practice games and no particular feelings about second place.
3
70
we'd better not be in this timeline: cumulative p(doom)=0.7%, and yet...
1
8
498
on the flip side, @chris_kraeuter and I reached 77.7% chance of cumulative p(doom), and yet we walked away with billions the gods giveth and the gods taketh away
1
8
159
ok, this is really fun! highly recommend taking a shot at it
1
7
970
On my way to @ethconf, too many good people going to be there to miss it If you’re in town and want to meet up, lmk
1
2
34
1,949
After 7 incredible years, I've decided that Friday May 29th will be my last day at the Ethereum Foundation. I'm humbled by the projects I got to work on along the way: from the KZG ceremony, to helping architect the early design of the Beacon Chain, and a lot in between. At the age of 23, the Ethereum space welcomed me on the basis of having some great (and many stupid) ideas and let me influence a multi-billion dollar technology, an incredible opportunity I will remain forever grateful for. Ethereum has had a huge impact on me, and I hope my work has had an impact on Ethereum, and in turn on the world. To every researcher, core dev, EFer, and community member, whether we worked together closely or not: thank you. The strength of Ethereum is, and always will be, the people behind it striving to make it what it is. I'm grateful to have spent these years among you. What's next: I don't entirely know yet. For now I'll be enjoying time with my wife and our 1-month-old while I figure it out. Longer term, I'll find or create something with brilliant people at the intersection of engineering excellence, hard problems, and useful products driving economic activity at scale. If that resonates, or you just want to catch up, slide into my DMs!
60
28
15
491
139,192
carlbeek retweeted
An exciting initiative is launching: The push for shorter slot times in Ethereum! The idea: Instead of a one-time change, parameterize the slot time and treat it similar to the gas limit - continuous performance improvements over time. And again a beautiful website!
the QuickSlots ⚡️🎰 website is live, and I'm very proud of the design! take a look! quickslots.info/ QuickSlots is the pragmatic way to get Ethereum to run much faster
18
23
1
169
21,623
carlbeek retweeted
ethereum goes faster
the QuickSlots ⚡️🎰 website is live, and I'm very proud of the design! take a look! quickslots.info/ QuickSlots is the pragmatic way to get Ethereum to run much faster
1
2
18
2,307
Ethereum, just faster.
the QuickSlots ⚡️🎰 website is live, and I'm very proud of the design! take a look! quickslots.info/ QuickSlots is the pragmatic way to get Ethereum to run much faster
1
2
1
25
2,198
the QuickSlots ⚡️🎰 website is live, and I'm very proud of the design! take a look! quickslots.info/ QuickSlots is the pragmatic way to get Ethereum to run much faster
13
35
12
143
45,083
carlbeek retweeted
personally i think we should optimize for area under the curve of CROPS delivered and number of people
1
1
5
652
Being evil is cringe
It takes time to unpack the noise, but @CharlotteFang77 's "it's just art" excuse is the contemporary equivalent of the tasteless YouTuber "it's just a prank bro" defense of plain bad parasitic memes Being evil is cringe
4
3
36
2,817
Ethereum ♥ money Ethereum ♥ markets Ethereum ♥ finance
[x] Ethereum is for money, markets, and finance.
3
3
33
1,740
right. open access isn't just a nice principle, it's literally just a better way to allocate capital
One aspect of DeFi that has been getting overlooked is efficiency Yes, there's trustlessness, transparency... but fundamentally a system where anyone can lend and anyone can provide liquidity is much more efficient than one where that is gated
10
850