Max

453 posts

Max banner
Max

Max

@0xM4x

TradFi backend engineer | web3 infrastructure | investor

on-chain Katılım Aralık 2016
372 Takip Edilen131 Takipçiler
Max
Max@0xM4x·
@Teknium cron jobs were bloated with tools they never used scoped it down, ~60-70% fewer tokens per run, merged in v0.11.0 excited to keep contributing👀
English
0
0
1
129
Teknium 🪽
Teknium 🪽@Teknium·
Introducing Hermes Agent v0.11.0 Our largest update yet, with over 700 PRs across ~200 contributors. Thank you to everyone who's worked on Hermes Agent! This update features a beta TUI v2, unlimited recursion depth and width of subagents, 5 new LLM providers, expanded image gen providers, QQBot gateway channel, themes & plugins for the dashboard, and so much more. Check out the main post below or see the release notes: github.com/NousResearch/h…
Teknium 🪽 tweet media
Nous Research@NousResearch

Hermes Agent v0.11.0 - “The Interface Release” Full changelog below ↓

English
77
111
1.5K
143.7K
Max
Max@0xM4x·
@0xuberM wanna build smth? dm me
English
0
0
0
23
Max
Max@0xM4x·
@izebel_eth this is why HL outshipped everyone. most protocols have a product team interpreting user feedback through jira tickets, HL has a founder who hits the same bugs before you do and patches them the same day
English
1
0
1
451
jez (equity perps era)
jez (equity perps era)@izebel_eth·
reading the hl feature was peak nostalgia, ive got stories like you wouldnt believe so many before hl, then hl redefined so much. had pnl swings that would make me sick now, and been saved more than once by luck/instinct/good advice all to say: be your own best user, and own the best things you use
jez (equity perps era) tweet media
jez (equity perps era)@izebel_eth

jeffs rapid response to trader asks has always been what i point to when ppl ask me "what makes hyperliquid different" its hard to describe why the ux feels good, but imagine how many times jeff has drilled into issues like this and made improvements

English
9
2
189
24.9K
Max
Max@0xM4x·
the label is "binary options" but the real play is that outcome markets are built into hypercore as a primitive. same execution environment as perps, sub-second matching, market makers can atomically hedge perp positions with binary outcomes on the same engine. that's not a new product - it's a new composability layer for risk
English
0
0
1
702
Arthur Hayes
Arthur Hayes@CryptoHayes·
HIP4 enables binary options on @HyperliquidX I expect an explosion of volume. $HYPE = $150 is just the beginning.
English
109
69
857
102.9K
Max
Max@0xM4x·
@Austin_Federa fox and cnn paid professionals to manufacture outrage on a schedule. X made it permissionless and 24/7. turns out the bottleneck to monetizing polarization was never demand - it was supply
English
0
0
0
46
Austin Federa | 🇺🇸
Austin Federa | 🇺🇸@Austin_Federa·
X has successfully monetized polarization to an extent fox and cnn could only dream of
Austin Federa | 🇺🇸 tweet media
English
8
2
46
3.5K
Max
Max@0xM4x·
@lopp 7.7tb across 337k files averages 22mb per file. this is raw unredacted bodycam and interview footage hitting the wire not just case text
English
0
0
8
418
Max
Max@0xM4x·
this is the AWS playbook. amazon ran AWS at near-zero margin for years, reinvested everything into infra, and by the time competitors showed up the moat was already built. if base hits escape velocity before L2 margins compress, the ecosystem lock-in becomes the margin. the bet is that you get there first
English
1
0
1
69
jesse.base.eth
jesse.base.eth@jessepollak·
tl;dr on how I think about @baseapp fees
jesse.base.eth@jessepollak

100% of revenue (+ a bunch more investment) from those fees goes back into growing the @baseapp and @base more broadly. this new referral system is one example of that, but every other program we run, product we build, and investment we make to grow is also funded by this (and the revenue from @base chain). our general philosophy is that we don't want to be the cheapest place to trade. we want to be the easiest and most trusted — and we want to build a flywheel that grows the overall economy. we have a lot of work to do to deliver on those goals but that's how we see it. I hope we can earn your usage of the app because you love using it. and I hope that our actions and impact demonstrate that the 1% fee goes towards the highest ROI things to help the base ecosystem (and all of the people who make it what it is) win. if you're curious about the broader way base operates inside of Coinbase, shared some more thoughts below. thanks for giving the feedback - hope this is useful context!

English
36
17
175
12.9K
Assasin
Assasin@assasin_eth·
Heyy can you guys recommend some books to learn, trying to strengthen things outside of crypto
English
80
0
152
8.8K
Max
Max@0xM4x·
ETH can't take shortcuts because no single party can force them through. that's also why quantum prep is happening years early: there's no CEO to say "ship it later, we have a roadmap" the same property that makes ETH slow is the one that makes it the only chain you can trust on a 20 year horizon
English
0
0
0
88
Uttam
Uttam@uttam_singhk·
ok ngl I was pretty bearish on ETH at the end of last year & was honestly close to selling everything but I am feeling pretty bullish again - I think it’s still the only chain that actually feels aligned with the whole cypherpunk/decentralization vision while other chains are prioritizing institutional adoption (which I don’t think is wrong btw) and now with the whole quantum threat + EF’s proactiveness & being ahead of the curve ETH seems better prepared than most although ETH flipping BTC is not gonna happen lol, don’t listen to influencers
English
29
10
167
8.7K
Max
Max@0xM4x·
if AI can reason about code well enough to find zero-days humans missed for decades, why would you still write formal proofs? the thesis is "AI makes verification cheaper." the counter-thesis is that AI makes verification unnecessary. you don't need to mathematically prove your code is correct if an agent can just audit it continuously and find what's wrong faster than you can write the spec
English
0
0
0
264
Georgios Konstantopoulos
Georgios Konstantopoulos@gakonst·
I feel similarly. We maintain Foundry, which is a fuzzer cosplaying as a day-to-day build tool for EVM smart contracts, primarily used in finance code. Over $100B has been deployed with Foundry. AI means we'll be writing 100x better properties with Foundry. This is very similar to Kleppmann's thesis on AI & the growth of formal verification in the future.
Georgios Konstantopoulos tweet media
Kenton Varda@KentonVarda

Honestly "AI that can find every vulnerability" sounds way better for the good guys than the bad guys. Not sure why everyone is losing their minds here.

English
13
5
133
23.5K
Max
Max@0xM4x·
Sui and Aptos built better VMs with better languages: safer type systems, resource model, no reentrancy by design. real technical wins. but they also asked every dev to abandon their tooling and start over. Monad kept the EVM and just made it fast. result: @Uniswap, @CurveFinance, @Balancer already live. @aave voting to deploy. $355m TVL in months. better tech doesn't win if nobody migrates
Max tweet media
English
0
0
0
57
Max
Max@0xM4x·
@jessepollak @baseapp the UX is great, but tokenized equity doesn't solve the oracle problem. bridging offchain legal reality to an onchain token still requires trusted middlemen. so what middlemen did we actually remove?
English
0
0
0
5
Max
Max@0xM4x·
@jessepollak people want this until they see the technical implementation. mandating protocol-level compliance or forcing sequencers into MSB roles will just push all actual innovation back to anon teams
English
0
0
1
139
Max
Max@0xM4x·
@mert 8-12 months is exactly when infra actually ships. speculators hate it, but for devs, the noise filtering out makes it the most productive window. quiet markets = pure focus
English
0
0
0
306
mert
mert@mert·
crypto if your time horizon is on the order of 8-12 months, it's a bad time to be here if your time horizon is anything greater than that, then it's a great time to be here
English
123
38
730
52.7K
Max
Max@0xM4x·
none of this is wrong. the list is real. but listing what crypto could do is not the same as explaining why it hasn't after 15 years and hundreds of billions in funding. the bottleneck was never blockspace. it's that every cycle we rebuild infra for the same users trading the same tokens
English
0
0
1
97
mert
mert@mert·
enough with the emotional self-defeatist mentality in this industry yes. there's tremendous potential left in crypto. we haven't even surfaced 1% of it crypto is the universal API for money, markets, and capitalism blockchains didn't even scale past steam-engine performance until 2 years ago unstoppable private money, programmable finance, internet capital markets, perpification of all assets on earth, decentralized coordination for physical infrastructure, lightspeed planetary payments with a line of code, a sanctuary economical system for anyone who didn't luck into being born in the west, zero knowledge proofs, fully homomorphic encryption, freedom, self-custody of assets without requiring permission from suits, property rights, contract enforcement, capitalism based on cryptography instead of the crony trillions
Artem Chystiakov@Arvolear

Is there anything exciting left in crypto?

English
57
67
557
32.3K
Max
Max@0xM4x·
survivorship bias in real time. you see 10 scammers who made it, not the 500 who launched the same token and got zero traction, zero exit, zero reputation left. patience doesn't feel like an edge until the cycle turns and you're the only one still standing with a clean name and actual skills
English
0
0
0
217
Uttam
Uttam@uttam_singhk·
discuss
Uttam tweet media
English
89
7
382
29.8K
Max
Max@0xM4x·
@lopp legacy media keeps hunting for a physical body to attack over an architecture they cannot kill
English
0
0
0
93
Jameson Lopp
Jameson Lopp@lopp·
The humorous thing about every Satoshi "reveal" is that you know, somewhere out there, Faketoshis are seething that someone else is getting the attention.
English
18
9
135
7.9K
Max
Max@0xM4x·
@lopp a secret held tightly enough without peer validation becomes indistinguishable from a hallucination
English
0
0
4
1.1K
Max
Max@0xM4x·
@pmarca offense scales linearly with compute while defense still stalls on human consensus. generating the patch doesn't magically enforce network migration.
English
0
0
0
90
Marc Andreessen 🇺🇸
Marc Andreessen 🇺🇸@pmarca·
The state of cybersecurity has been dismal forever. At one point a major vendor even enabled direct execution of arbitrary x86 binaries in any web page. Nobody cared. The number of hacks and breaches has been uncounted. Finally we have the catalyst and the tools to fix it all.
Ethan Mollick@emollick

Curious how many large organization CISO offices have taken the Mythos red team reports as the red alert that it is. (I suspect very few) Based on historical trends in AI they have, at most, about six to nine months until those capabilities become widely diffused to bad actors.

English
105
109
1.3K
170.3K

Keşfet

@Teknium @0xuberM @izebel_eth @HyperliquidX @Austin_Federa @lopp @baseapp @assasin_eth