kaereste.gwei 💗🦇🔊

4.6K posts

kaereste.gwei 💗🦇🔊

kaereste.gwei 💗🦇🔊

@kaereste

Krzysztof Urbański, governance and partnerships @L2BEAT "Misty Drag is a Moonshot Bot that builds coordination for public good."

Warsaw Katılım Nisan 2007
4.3K Takip Edilen2.4K Takipçiler
binji
binji@binji_x·
46,000 followers today but I miss my old friends, I barely see you guys anymore wya.
English
50
0
150
9.7K
kaereste.gwei 💗🦇🔊
Crypto has a BD prob^W opportunity. Do a simple experiment - ask five random normie friends what they know about crypto and DeFi. Most likely, the answer you’ll get will be some variation of “buy Bitcoin low, sell high”. If they’re a bit closer to finance, they may also say something about stablecoins for payments, the difference between “crypto” and “digital assets”, and how “it” is the future, but if asked further they won’t be able to give many examples. If you ask them whether they know Aave, Morpho, Spark or even Uniswap, usually the best you’ll get is that they’ve heard the name but have never used it themselves. I’ve done this many times, sometimes, surprisingly, in groups that technically “should” know better because they are directly involved in work related to digital asset strategies for institutions. Not to mention that for many of them, the mainstream narrative of the last few years has been a steady stream of scams, frauds, collapses, controversies, and terminally online drama. Some may see this as a problem. I see it as an opportunity. Because it’s pretty easily addressable. Not by another high-level deck. Not by another panel where everyone already agrees with each other. You address it by going into the rooms where these people actually are and showing them what people in crypto take for granted. Some years ago I did several workshops for people in traditional finance - lawyers, managers, analysts, compliance people, traders, but also some technical folks - going with them through @Speedethereum challenges, which back then was basically scaffold-eth plus playing with some burner wallets. They weren’t really writing any code. I was giving them snippets to copy-paste into Remix, but the code was short enough that they were able to grasp what was happening. You could literally see lightbulbs over people’s heads when they traded their custom-made ERC-20 tokens through a permissionless escrow. Crypto projects need to do more of that. Much more. You need to acknowledge that there is a word “development” in business development. It’s not just sales. People need to know what you’re selling. The decision-maker is not the only audience. Their colleagues matter. Their lawyers matter. Their risk people matter. Their friends in other institutions matter. You want the person choosing your solution to look less like someone betting their career on weird unknown tech and more like someone experimenting with a new infrastructure category that serious people are already learning about. That means meeting people where they are. I love Devcon and Token2049 as much as anyone. But if you want finance, fintech, payments, compliance, and institutional people to understand what you are building, maybe you also need to be at places like Singapore FinTech Festival. People who have been in crypto for years forget how much they had to digest along the way. People arriving this year do not get that context by osmosis. The industry should not outsource all of the hard work to them. Crypto is already good enough from the tech side, now let's improve the BD. I am putting together several hands-on workshops for non-crypto audiences this year. What should I show them?
English
2
0
11
381
kaereste.gwei 💗🦇🔊 retweetledi
donnoh.gwei 💗
donnoh.gwei 💗@donnoh_eth·
PSA: we (@l2beat) just removed ~$7B of $RAIN non-circulating supply from @arbitrum TVS held in a few multisigs controlled by the team itself. If you see Arbitrum TVS being significantly lower it's because of this change, nothing major happened.
English
3
4
42
4.4K
kaereste.gwei 💗🦇🔊 retweetledi
L2BEAT 💗
L2BEAT 💗@l2beat·
We are expanding our Interoperability with a new Intent Bridges dashboard, a key tool to understand how liquidity moves fast and efficiently across the ecosystem. What does the intents space look like today? Let’s dive into the data
L2BEAT 💗 tweet media
English
2
9
44
3.4K
Ben Greenberg
Ben Greenberg@hummusonrails·
Somehow I ended up in a pub in London and there’s a soccer game happening. I think it’s a big one. 🏴󠁧󠁢󠁥󠁮󠁧󠁿🇳🇴
Ben Greenberg tweet media
English
3
0
19
1.1K
Andy
Andy@andyyy·
Turned 27 today. 26 was one of the most challenging, yet rewarding years of my life 🙏
Andy tweet mediaAndy tweet mediaAndy tweet mediaAndy tweet media
English
246
5
726
78.9K
kaereste.gwei 💗🦇🔊
FSD is inevitable. The sooner we embrace it and start redesigning our infrastructure for it, the better. I've lost a couple of friends in car accidents. I hope my kids will live in a world where manual driving happens only in specially desginated areas.
@levelsio@levelsio

🇵🇹 Carta aberta ao Governo de Portugal @LMontenegroPSD @Leitao_Amaro @miguelluz Senhor Primeiro-Ministro Luís Montenegro, Senhor Ministro António Leitão Amaro, Senhor Ministro Miguel Pinto Luz, Escrevo como alguém que escolheu Portugal para viver. Tenho um pedido simples: tragam o Tesla FSD (Supervised) para Portugal. A 10 de abril, os Países Baixos foram o primeiro país europeu a aprovar o sistema, depois de 18 meses de testes independentes da autoridade RDW. Em dois meses, a Lituânia, a Estónia, a Dinamarca e a Bélgica seguiram o exemplo. Nenhum destes países repetiu os testes. Reconheceram a homologação neerlandesa ao abrigo do quadro europeu (UN R-171) e aprovaram com base nos dados que já existem. Os resultados falam por si: nos Países Baixos, cerca de 40.000 Teslas já percorreram 24 milhões de quilómetros sem incidentes relevantes, segundo a própria RDW. Os carros com FSD registaram 3,5 vezes menos colisões do que a condução manual. E é aqui que isto deixa de ser sobre tecnologia e passa a ser sobre vidas. Só este ano já morreram mais de 220 pessoas nas estradas portuguesas, mais 25% do que no ano passado. Portugal está acima da média europeia na mortalidade rodoviária. A ANSR diz que as principais causas são humanas: velocidade, álcool, distração. Um sistema que nunca se distrai, nunca bebe e nunca adormece evita mortes que vão acontecer com condutores humanos. O próprio Governo chamou à sinistralidade uma chaga social. Esta é uma forma concreta de a combater. Portugal pode fazer o mesmo que a Dinamarca e a Bélgica: o IMT analisa o dossier da RDW e aprova. Não custa nada ao Estado e coloca Portugal à frente da Alemanha, da França e da Espanha. Há 500 anos, Portugal liderou o mundo numa nova tecnologia: a navegação. Partiu à frente de todos e deu novos mundos ao mundo. Portugal pode voltar a estar na fronteira de uma nova tecnologia e liderar como fez há 500 anos. Cada mês de espera custa vidas. Basta decidir. Com admiração e respeito, -@levelsio

English
2
0
12
1.2K
kaereste.gwei 💗🦇🔊
“When you keep pushing your agents, you end up in a place where no soul has been before. The mathematics is completely foreign, and the notation is concocted and unfamiliar. Your instinct is to say that it is all wrong, but it isn't. It is just alien.”
Bartosz Naskręcki@nasqret

I've been doing a lot of experiments with auto-research in the last few weeks, especially in algebra. Here are a couple of thoughts I want to share about sparks of generalization. 1. Once the problem can be reduced to a constructive question about rings and becomes a finite computation, I can use a model to basically develop the code from scratch. You don't even need a specialized CAS for that, Rust is enough. 2. If the question is about a highly abstract object, you can still turn it into a format in which all the constraints can be manipulated formally. You can use functional programming to model the abstraction through constraints. The best situation is when you can actually encode the abstraction in one of the many logics that are friendly to SMT solvers. 3. Most of the time, the difficulty of the problem lies in the lack of clear connections between some of the encoded predicates. This is where it starts getting interesting. When you play with code generation, you can accidentally discover such missing rules and abstract them from the calculations. 4. Here, there is a chance that the model will spontaneously generalize the pattern into an abstraction. This is, in general, very hard because the models are not designed to produce something unusual. Quite the contrary: they usually do not get very far beyond their training. 5. This is where something strange must happen with the model: an accidental guess, multi-agent search, etc. I have already seen some very small sparks of new ideas or accidental generalization in my auto-research loops. I suspect this will scale up with compute. 6. Now the difficult part begins for humans: those abstracted rules might be provable, or even formalizable in Lean, but they are completely new and foreign to humans. I have seen this in my research into invertible cellular automata. I see a claim that is true—there is a proof—but I don't understand its deeper meaning. I don't know how to internalize it off the bat. 7. When you keep pushing your agents, you end up in a place where no soul has been before. The mathematics is completely foreign, and the notation is concocted and unfamiliar. Your instinct is to say that it is all wrong, but it isn't. It is just alien. In the coming days, I will publish several such examples straight from my loops. I suspect this is an early instance of spontaneous and very weak generalization. It feels like the early hacks with GPT-3.5 that produced something resembling “thinking.” It is very bad, but it is something. Maybe ideation is mechanical and can be scaled up with a lot of effort? Where are we going?

English
3
0
7
632
kaereste.gwei 💗🦇🔊 retweetledi
Octant
Octant@OctantApp·
While privacy policy takes a few steps backwards with Chat Control, our grantees @l2beat are making it more accessible with their new dashboard. A censorship-resistant, open-source, free, private and secure world should be accessible to everyone.
L2BEAT 💗@l2beat

We built our Privacy Dashboard to make some of the CROPS principles easier to inspect and compare across privacy protocols. Here's how each column on the page maps to the principles 👇

English
0
2
13
773
kaereste.gwei 💗🦇🔊
This is remarkable. The positive news here is that access to a genius-level algorithmic expert has now been officially commoditised and made abundant. I’m curious what my life would have looked like if this had been accessible when I was 15. Tinkering with problems like this was a big part of my life back then, but I had literally only a handful of friends around me I could talk to about it. Now any frontier LLM can be a friendly companion you can discuss this stuff with all day, in detail. And congrats to @FakePsyho. I’m quite convinced you will forever remain the last human to beat AI at competitive problem solving.
Psyho@FakePsyho

Quick summary of what happened during AWTF in Japan: - OpenAI crushed humans in both Heuristic and Algorithm categories. Heuristic is the one where I won in 2025 and OpenAI placed 2nd - AWTF is pretty much the highest level competitive programming contest: invited finalists in both categories are among the best people in the world; this is essentially the first time AI won vs humans in a programming competition in such a decisive matter - big change compared to 2025 is that AI was constantly progressing and didn't get stuck / plateau at any point - System used for Algorithm is mostly just custom harness + model very close to gpt 5.6; my educated guess is that system used for Heuristic is a custom "autoresearch" harness with a swarm of cooperating agents and a massive inference cost - OpenAI was sure that they will win Algorithm and were very confident that they will win Heuristic; they did a lot of backtesting on old contests and they would win all of them in their simulations - imho heuristic problems are a great proxy for ML autoresearch capabilities; if AI was able to match best humans here, we're very close to RSI / automated researcher; this result is way bigger than a high score on some questionable benchmark When I get some rest, I'll try to post more thoughts about the whole thing. Thanks for following the coverage. I believe I tweeted more in those few days than I did in the first half of 2026. And I'm not even counting the 10 hour long livestream. Oh and technically speaking, I didn't participate so I'm still undefeated and I'll gladly keep "Humanity's Last Programmer" in my bio.

English
0
0
6
635
kaereste.gwei 💗🦇🔊 retweetledi
autist
autist@litteralyme0·
Growing up hearing “Wikipedia isn't a valid source” and then entering a workplace where people say “just ask ChatGPT” is a surprisingly strange timeline
English
627
87.2K
633.6K
6.6M
kaereste.gwei 💗🦇🔊
@maxlomu no need to revisit anything, it’s the current state just with extra overhead. And it doesn’t have to be necessarily bad state, details matter. There are other questions worth asking though, yet probably not via public X threads.
English
1
0
1
166
maxlomu 🤌
maxlomu 🤌@maxlomu·
@kaereste Last occasion IMO to prove the DAO can achieve meaningful speed. If by the end of this cycle we still see the slow pace and limited results of the last period, it’s worth revisiting the whole concept and whether we should just hand everything to OCL and the Foundation
English
1
0
3
149
kaereste.gwei 💗🦇🔊
ArbitrumDAO is having an election for the OAT Council, and I think it deserves more attention than it has received so far. So I want to share a few personal thoughts on the matter. A disclaimer: these are my personal views only. We have not yet discussed in the governance team how we will vote. Why is this election important? For two reasons. 1. this committee is, in practice, overseeing and driving not only OpCo, but also many cross-ecosystem initiatives, and 2. it is the last avenue in ArbitrumDAO where tokenholders, via delegates, can directly express their views on how they want the DAO to evolve. As a long-time ArbitrumDAO community member, I want to use this opportunity to share my own view. The tl;dr is that while I believe the overall direction is correct and promising, I am disappointed with the execution style and performance, and I am concerned about the state of the DAO today. At the same time, I am still hopeful. I believe the future can be bright, but some things need to change. First, I want to say that I have huge respect for last year’s committee members. I believe they are highly competent individuals, and I stand by our election votes from last year. But this year, I would not vote for the same people. Not because I don’t believe they are individually suited for the role. But last year proved, in my opinion, that collectively, the OAT did not work as (at least I) expected. Instead of becoming more than the sum of its individual parts, it became something like the safest common denominator. The result was that the OAT and the broader DAO significantly disappointed and underperformed. I will repeat, though, that I still believe in the overall vision formulated at the beginning of last year, when OpCo and the OAT were formed. I personally supported it, and I still believe it is the right path. But I also have to honestly admit that many of the risks and dangers raised by critics back then materialized, while many of the benefits and advantages I was looking for did not. One of the features of having an OAT is that when results are disappointing, there is someone to point to and a clear place to call for change. I’m not necessarily saying that the current developments are bad, although I do have concerns here and there. I am much more concerned about the things that are not happening, and about the pace of development. Anyone who reads the proposals and related forum commentary from last year will see that the DAO, OpCo, and the OAT are nowhere near where they were supposed to be. By my personal expectations, maybe 20% was delivered, and most opportunities were missed. Some things have not been delivered at all. Many pivots may have been justified, but they were never approved by, or even consulted with, the DAO. The DAO itself became stale and almost completely inactive, to some extent by design and on purpose. Not to mention that the token price is far from tokenholders’ expectations. The overall market can be used as an excuse, but I would say that it is just that: an excuse. I don’t see the DAO doing anything meaningful to counter that trend. On the contrary, I see an entity vs. tokenholder conflict, similar to what we have seen in Uniswap, Aave, and ENS, looming on the horizon. The OAT/AAE vision promised better governance. So far, it has delivered governance in a coma. ArbitrumDAO used to be known for a great community and a welcoming DeFi ecosystem. A place to experiment, push boundaries, and bend reality. ArbitrumDAO used to experiment with ways to support builders. ArbitrumDAO created an RWA investment program (STEP) which attracted major financial institutions and supports Arbitrum’s budget today. ArbitrumDAO created a real investment fund. ArbitrumDAO funded incentive programs that attracted 100+ DeFi projects. Last but not least, ArbitrumDAO created OpCo, the OAT, and the rest of the foundations for what we have today. Of course, not everything the DAO did in the past was great. The changes we introduced were introduced for a reason. But I feel the pendulum swung way too far in the other direction. I probably don’t have a complete enough overview, but my impression is that over the past year, ArbitrumDAO has not produced anything materially significant. There is barely any meaningful discussion happening either. And it’s not as if the current state of affairs is clearly better than what we had before. I like some of the new things, but I miss too many of the things that are now gone. Most importantly, we no longer have any real DAO oversight or control. There are no meaningful feedback loops, just back-scratching and institutional inertia. ArbitrumDAO used to be Pied Piper. Now it feels like Dunder Mifflin Paper Company, the one just before the merger with Sabre. So what would I like to change? Def not everything. The overall direction set by the previous OAT should be maintained, but executed better. I want to go back to the original spirit of governance advancement. Creating meaningful governance, not waiting for it to magically appear out of nowhere, and definitely not restraining it (inb4 later complain that governance is not working and should be removed). More communication, more openness to new ideas, less reliance on single vision-drivers. I want Arbitrum to strive to become a business greater than its creators, not the other way around. Working constructively with other entities, but as equals, not subordinates. But most importantly, I want DAO delegates to discuss this, state their views, and actively engage. Top delegates will choose the committee that will drive ArbitrumDAO’s future. I hope each of them does their research, formulates their expectations, and votes with confidence that they are choosing the people most aligned with those expectations. Not just going with the default to get it off their plate. I’m pretty sure Arbitrum will be okay no matter how the election settles, but DAO delegates should strive for it to be great, not just okay. Finally, I want to address the inevitable comments that I am criticizing and calling for change only because I have been sidelined and am not as actively involved as I used to be. These are fair arguments. And I won’t lie, I do miss, a bit, the times when I spent whole days on DAO-related calls. But I am genuinely fine doing what I am doing now. I have more ideas in my head than I will ever be able to tackle, and I am perfectly fine with others pushing the cart today. I just want the cart to go in the right direction. ArbitrumDAO’s fate is important for the future of DAOs, and Arbitrum’s fate is important for the future of crypto. I want both of those futures to be bright. P.S.: Of course, the fact that I'm writing this here on X instead of the forum is a statement in itself, IYKYK.
kaereste.gwei 💗🦇🔊 tweet media
English
7
6
35
2.4K
kaereste.gwei 💗🦇🔊
Robinhood Chain launch is a great success and a big win for Ethereum and Arbitrum. Period. I thought this was obvious, but I realised - thanks to feedback from a fellow community member, which I appreciate - that I hadn’t said it out loud. You may have seen me lately sharing critical thoughts about ArbitrumDAO and Arbitrum. Those criticisms stand, but they were about governance execution, missed potential, and the expectations I have as someone who has been close to the ecosystem for a long time. They were not criticism of the Robinhood Chain launch or of the Arbitrum stack. So I want to make one thing clear - Arbitrum is a great project. Robinhood choosing the Arbitrum stack and successfully integrating it into its product line is a major validation. It shows that Arbitrum is production-ready infrastructure that can deliver real business advantages for fintechs. I’m already sharing the Arbitrum/Robinhood case in my TradFi circles as a success story and a pattern to follow. Every TradFi company should look at this case for inspiration on how to think about its crypto strategy. I criticize ArbitrumDAO precisely because I think Arbitrum has huge potential. I care because there is something real here. My expectations for Arbitrum are high. I don’t spend this kind of energy on projects I think are irrelevant.
English
2
2
71
3.4K
kaereste.gwei 💗🦇🔊
@hypermaddd probably bc building an Ethereum L2 is a real thing with clear guidance as how to do it properly via l2beat.com and “Solana subnet or something” is a thing you made up as sth that should be possible but no-one tried it yet for real
English
2
0
10
250
Hyper
Hyper@hypermaddd·
Why Robinhood choice Arbitrum instead of solana subnet something?
English
4
1
4
2.3K
kaereste.gwei 💗🦇🔊
Don’t want to spoil all the fun for the recent narrative spin but business license is not the type of moat some of you hope it to be. Esp when it comes to software you run yourself, not share. Ask anyone selling software in the late 90s/early 2000s. But I guess we need to repeat the history, make no mistakes.
English
2
0
10
890