Zac

45 posts

Zac

@zholme7

building blocks @sigp_io https://t.co/sLFTC7pxWO

Chicago, IL Katılım Ocak 2026

341 Takip Edilen66 Takipçiler

Zac@zholme7·2d

all protocols that fork uniswap just to change the callback name should go straight to zero

English

Zac@zholme7·9 May

@0xQuintus not yet but was inspired by this talk to give some solana pAMMs a go m.youtube.com/watch?v=_VPQHv…

English

632

quintus (sovs arc)@0xQuintus·9 May

@zholme7 Have you tried much non-evm reverse engineering?

English

425

Zac@zholme7·9 May

no smart contract is safe from mevlog, heimdall, cast, and gpt5.5 xhigh. have fully reconstructed a few really cool ones

English

6.1K

Zac@zholme7·27 Nis

@GiulioRebuffo @OpenAI got the same flag seems like they’re being very precautious

English

1.9K

Giulio Rebuffo@GiulioRebuffo·26 Nis

dear @OpenAI, I am not trying to hack anyone. I am just trying to test increasing Ethereum's contract size to 64KB. this is a matter of uttermost urgency

English

141

17.9K

Zac@zholme7·25 Nis

@lublu_mev cool, makes sense

English

lublu mev@lublu_mev·25 Nis

@zholme7 1. He’s doing this against reorgs. 2. msg.value is cheaper than passing through calldata.

English

170

Zac@zholme7·24 Nis

instead of just setting block number on the bundle this mev bot also has a modifier that will check during execution that msg.value == block.number, interesting etherscan.io/address/0xbdb3…

English

345

Zac@zholme7·24 Nis

@mempirate haha knew that was gonna be called out

English

mempirate@mempirate·24 Nis

@zholme7 Up that swap

English

Zac@zholme7·24 Nis

sidequests have outgrown server right when storage and ram prices are at the peak..

English

257

Zac@zholme7·20 Nis

I regret taking low gas prices for granted

English

162

Zac@zholme7·19 Nis

rsETH finally convinced me to unwind all positions and hole up. Yield on chain right now does not justify constant impending doom

English

164

Zac retweetledi

Blockspace Forum@blockspaceforum·16 Nis

1/ In Cannes, Blockspace Forum held its second workshop, attended by 32 teams representing >95% of out-of-protocol blocks. We’ve summarized the key outcomes and concepts that surfaced. Thank you to everyone who contributed. The ticker is ETH. 🧵👇 blockspace.forum/learn/events/c…

English

7.9K

Zac@zholme7·16 Nis

Maybe a bit LLM ungrateful but kind of a disappointing drop. They're riding the mythos high and this is a perfect time for OpenAI to drop a consumer available model that rivals mythos. Even with 4.7 I dont have any desire to switch from 5.4 & 5.5 will probably really lock me in.

Claude@claudeai

Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision.

English

206

Zac@zholme7·9 Nis

@BowTiedDevil appreciate it!

English

BowTiedDevil@BowTiedDevil·9 Nis

@zholme7 This is good stuff

English

Zac@zholme7·9 Nis

Just put the cycles in the bag bro. Quick writeup about benchmarking with rdtsc & rdtscp thinkingthoughts.dev/blog/benchmark…

English

538

Zac@zholme7·7 Nis

firedancer code documentation is seriously amazing. goldmine for learning some super cool techniques

English

174

Zac@zholme7·6 Nis

cool way I just discovered to precisely benchmark part of your code with perf(1). Setup a FIFO that you write `enable` & `disable` to in bracketed around your hotspot code