i look at pure research with a bit of contempt. it's for people who don't get their hands dirty and don't want to take the risks of seeing their idea perform in production. you just spent years in an insulated ivory tower, while others have a live feedback loop collecting data in the trenches.

0/ Clear signing is now live. An open standard to end blind signing, making human-readable transactions default. This effort brings a major UX and Security upgrade to transaction signing on Ethereum.

EIP-7702 is the greatest marketing disaster in Ethereum history... - no mainstream (or any that I know of) wallet implements a flow for signing an arbitrary delegation - attackers use EIP-7702 once they have the user's private key, to make draining easier - user sees EIP-7702 on chain and immediately thinks "I got phished into signing a delegation", not "my private key is compromised" For anyone in this situation who doesn't believe me, just send funds to your compromised address on a DIFFERENT chain. You will see a NEW delegation appear, without your involvement, and your funds get drained.

🔐 New EIP-8250: Keyed Nonces for Frame Transactions 🔐 by @soispoke, @nero_eth, @lightclients and @VitalikButerin This replaces the single sender nonce with (nonce_key, nonce_seq), giving frame transactions independent replay domains. For privacy protocols, the key can be derived from a nullifier: concurrent withdrawals from a shared sender become possible, with inclusion atomically marking the nullifier spent. Target fork: Hegota Links below 👇

Yesterday, we wrapped up the Soldøgn interop: a week long core dev event focused on hardening Glamsterdam implementations to scale Ethereum securely ☀️ It was our most intense one yet. Teams used every hour of the midnight sun, ultimately converging on a 200M gas limit target after Glamsterdam, a more than 3x increase! Soldøgn also marked the end of my tenure at the EF & L1 R&D. As announced earlier this year, I'll be exploring frontier use cases for Ethereum. I could not have asked for a better way to wrap up the past 8 years: IMO this was our best interop yet. Thank you to everyone who made it so special ❤️‍🔥 I’ll be offline for the next month, then back in June, kicking things off at @EthConf! Please reach out then to chat about things that only Ethereum can make possible. Cheers 👋

Sustainability of those building dev tooling in crypto is and has always been a pretty big issue. Here is how we are solving it 👇 Traditionally the options are: 1. Big company ends up building their own tooling and opens it up for everyone (e.g. Golang by Google). This is generally quite good until the company decides it doesn't care about the space anymore. Too brittle for Ethereum. 2. Dev tooling is tied to a broader product (intel's C++ compiler that was heavily integrated into intel servers). This is a walled garden. Very good for intel users, doesn't help anyone else. 3. Dev tooling is sold commercially (SaaS). Given crypto is still nascent, expecting new teams to pay up is pretty bad. Also SaaS is dead with AI. 4. You don't get paid. You just do it because you love it. This is fine, now with LLMs, especially for small SDKs. But it doesn't work for large scale projects that require teams for constant upkeep. E.g. compilers, language, L2Beat etc. So wat do? And NO the answer isn't EF pays for these devs continuously. The @ethereum Foundation aims to withstand the 'walk away' test i.e. not have teams dependant on itself, especially for grants. Hence @nachortti has been working on 'Project Odin': work with each of large-scale, complex, heavily used and open-source project on Ethereum and getting them to sustainability. Shout-out to @vyperlang who will be doing formal verfication work to sustain the project! Auditing, building POCs or stand-alone products for enterprise, partnerships are relatively easy wins for this archetype of builders since they are generally experts and brilliant engineers in what they do. Next time: before asking for a grant, think how can Project Odin help you. Think alternate sources of generating revenue.

I am pumped to announce that, as of April, I have been working at @ethereum foundation as the EVM Dev Tooling Lead. Bittersweetly, this does mark the end of my 3 (!!) years at @aztecnetwork. Thank you to @jaosef @lisacuesta and many others for the amazing memories, offsites and the massive learning opportunity, as I pivoted from engineering to product management. I am excited for what Labs will do post Alpha launch and serving as an example for privacy products in crypto. There is a lot in store there. In my new role at the EF, I hold myself accountable for making it easier to build EVM apps. I plan to tackle: 👉 Fixing top developer pain points and working with @HardhatHQ @solidity_lang, foundry, @etherscan, other SDKs, tooling and even all core devs. And yes, I have a list of the pain points. 👉 Ensuring tooling stays up to date with latest forks (glamsterdam, hegota etc) 👉 Working with the one and only @austingriffith and his team to ensure Claude code and other LLMs can build good, secure dapps (checkout btw!) 👉 Outlining broader language and tooling vision, funding strategies with appropriate teams, 👉 Sunsetting not needed tools or finding other maintainers. I have serving in this role for just about a month now and there is so much cooking. You will be hearing many fun announcements and big frustrations being solved over the next few weeks. Solidity devs - if you ever encounter any problem/frustration with any tooling - do reach out. My DMs remain open! To the X algorithm - help me be the point of contact for solidity devs! I want to know your biggest issues when building on Ethereum or L2s. PS: Is it cringe to say evm/acc?

Quick update on two draft standards that are more interesting together than apart: ERC-8211 and EIP-8141. Both are in Draft, both are being designed with awareness of each other, and the pairing is worth paying attention to. EIP-8141 is native account abstraction as a new Ethereum transaction type. Validation, gas payment, and execution become first-class protocol primitives — no alt-mempool, no bundler network, no EntryPoint contract. What ERC-4337 does in user-space, 8141 does in-protocol. The practical upshot: custom validation per account (passkeys, multisig, session keys, post-quantum), any party paying any transaction in any token, atomic batching at the transaction level, and every existing EOA getting a default behaviour inside the new tx type with no migration. ERC-8211 is the encoding question. Inside any transaction — native or otherwise — something has to fill the execution slot. Today that's usually raw calldata or a custom router contract the dapp had to deploy and audit. 8211 replaces that with a structured, self-describing batch: runtime-resolved parameters, inline assertions, and predicate gates carried as data, not as a deployed helper contract. The flow a frontend dev wants to ship ("deposit, rebalance, stake, claim") becomes a TypeScript intent compiled to an encoded payload. No new Solidity per combination. This isn't theoretical. The pattern has been running in production for over a year as Biconomy's Modular Execution Environment, serving users across 4337, 7702, and wallet_sendCalls. ERC-8211 is the formalisation of that pattern as an open standard. The synergy in one line: 8211 answers what's in the transaction, 8141 answers how the transaction travels. 8211 alone is an excellent payload that only reaches users already on a specific AA stack. 8141 alone is universal transport, but apps still hand-roll routers and signing schemes, and wallet screens still render as opaque calldata. Both together is the first time Ethereum has a uniform way to carry structured programs to every account. A few things stop being dedicated infra projects once both ship: composing DeFi flows without a deployed router, MEV resistance via inline minimums and runtime-resolved parameters, single-signature multi-chain that's bridge-neutral, and wallet signing screens that show outcomes instead of calldata. On that last one — because an 8211 batch is self-describing, a wallet can parse its assertions and render something like "this will succeed only if you end up with at least 0.35 WETH and 3,383 aUSDC." The user signs the outcome. Works on hardware wallets with no internet access. For builders: if you're shipping multi-step flows today, write them as 8211 intents on whichever AA stack your users are on. When 8141 lands, the same code reaches every EOA — no rewrite, no migration. For users: the benefits are already available via MEE on 4337 / 7702 accounts — single-sig multi-chain, contract-free composition, inline MEV protection, readable signing. 8141 is what extends that surface to the EOA you already have. Neither standard is a new primitive or a faster chain. They're a cleaner split of responsibilities: a transport that doesn't care what it carries, and a payload that doesn't care which transport it rides. Worth tracking both.