@GrapeBaBai
iAccount based inUnited States
About this account
- Account based in
- United States
- Connected via
- United States App Store
Account-level information from X, not a live location or the device used for a specific post.
Ethereum developer, work for @lodestar_eth and @zeamETH with @ziglang, creator of @zen_ethereum and @optimism_java
Hubei
Joined March 2013
- Tweets4K
- Following3.4K
- Followers487
- Likes5K
One step closer to 4-8x faster Ethereum finality!
It took some time and lots of tokens, but we now have a formally verified proposal for a decoupled consensus protocol in I* (a future Ethereum upgrade)! Not yet a full spec (up next), but it includes all the key consensus-relevant details to become one.
Since Ethereum aspires to be live without most of the stake online, the protocol involves many more components than a normal BFT protocol, and its correctness involves much more than standard safety and liveness. Those nuanced properties are now verified!
What's more, I came away convinced that all protocol design will involve AI-assisted Formal Verification in the future, both for correctness and iteration speed.
The work wasn't limited to just:
Design the protocol -> Formally verify it
Instead, the loop became more like:
Design -> Formal Model -> Find exactly what breaks and why -> Redesign it.
For a fairly complicated protocol like this one, I think having the Lean model be part of the design loop played a big role in accelerating the process.
A future with agents paired with formal models is a superpower for Ethereum development, because they can then use those models to find exactly where an argument breaks down, formalize counterexamples, test proposed fixes, iterate on the protocol.
Many details that would slip under the radar when asking agents (and indeed, humans) can now be specified exactly and checked by the Lean kernel. This then forces agents to be more precise and lets them make verifiable progress on their own. It's been incredible to see this play out, seeing agents find gaps and propose protocol changes to fix them.
In other words, autoresearch can speed up protocol design, formal verification is here to stay, and Ethereum Finality will get faster.
Valen is a programming language that's aims to be not only fast and memory-safe, but also easy and flexible.
github.com/valen-lang/valen
Kai Chen retweeted
Replying to @amritwt
Stephen Toub is a different kinda legend!
I remember watching the man live code async/await logic from scratch along with Scott Hanselman in a livestream before the vibe coding era with no separate notes or copy pasting
youtu.be/R-z2Hv-7nxk?si=Yjko…
怎么申请呢
#Rustlang Job
【帮友找 Rust 开发】我这有一个巴黎 Rust 职位,全职和兼职均可。
可远程,你人在国内和欧洲均可。
要求你有 Rust 手工开发经验 + AI Coding 经验。并且你要有处理支付的经验。
名额有限,满足要求的来,薪资面谈,岗位要求👇
Kai Chen retweeted
Our v1.48.0 release is now out!! We just added flat-file support for data columns and have sped up our block process times. Check out the release for more info.
github.com/ChainSafe/lodesta…
Kai Chen retweeted
Hegota should be a small fork that ships fast.
The Lodestar team ranked the proposed consensus-layer EIPs. Here's where we think the priorities belong.
S: must ship
A: strongly support
B: support if the timeline allows
C: support with qualifications
Kai Chen retweeted
We are proud to announce our preferences for Hegota EIP inclusions.
Please find our thoughts here:
hackmd.io/@matthewkeil/HkcMq…
Kai Chen retweeted
After more than 3.5 years, I just archived zap in favor of baz! Bounded Async Zap! 🤩
Pure Zig. Explicit bounds. Streaming. Native Linux, macOS & Windows.
Powered by bounded/http, my HTTP/1.1 engine inspired by @tigerbeetledb’s #TigerStyle!
Still early days, but looking promising!
technologylab-ai.github.io/b…
Kai Chen retweeted
Lodestar-z v1.0.0 is live! 🎉
Up next: Zig live on mainnet.
📦 Grab the release: npmjs.com/package/@chainsafe…
Kai Chen retweeted
for anyone that wants to join the network, it's very simple by running @lodestar_eth using our `chainsafe/lodestar:next` image with the `--network plataberget` flag 🐻❄️
Introducing the Platåberget testnet! Its our Glamsterdam focused testnet to help the community test all the upcoming changes. Please get involved with the testing efforts!
blog.ethereum.org/2026/08/17…
Kai Chen retweeted
one interesting design space in state is how clients can store tries efficiently. Back in the early days of Geth, users had to manually execute specific commands to prune the database. Thats no longer necessary because Geth migrated to a new state format called the path-based state scheme (PBSS) in 2023.
in the PBT specifically, storing each trie node individually is highly inefficient. A better approach is to group related nodes and store them as a single database entry, rebuilding the subtrie on the fly when needed.
there’s still plenty of room for optimization in this area. I highly recommend exploring PBT! :)
EIP-8297 Partitioned Binary Trie (PBT) prototype in @NethermindEth.
Another step toward accelerating Ethereum’s Strawmap.
As AI advances, we should think further ahead than the next fork.
github.com/NethermindEth/net…
Kudos to @asdacap
I spent 14 days in China, and it was my favorite trip in a while.
It changed so much since I visited 11 years ago that it felt like a new country. Here is what I liked about China, and what I didn’t.
1. EV adoption is huge. 100% of scooters are electric. You can walk on busy streets and it’s surprisingly quiet! It made me enjoy cities again.
2. It’s so convenient. All restaurants let you scan a QR code to order and pay. You can order a taxi and drive 100 km for $30. The malls are so big you can spend an entire day there.
3. Value for money is top notch. Hotels are new, food is delicious, and yet everything is super affordable. Chinese people are really committed to doing their best, and you can feel this everywhere you go. It’s very humbling.
4. Chinese people are super friendly! We made friends everywhere we went: restaurants, hotels, taxis. I memorized Chinese sentences like 是免費的嗎? (“Is it free?”) and I would repeat them to outdoor vendors for fun. They would laugh and come slap me on the shoulder! I even got a free ice cream doing this ^^
5. Internet is a gamble. The Great Firewall blocks a lot of websites. You need a VPN and eSIM to access Western websites. And it’s not guaranteed to work, so it’s tricky to get deep work done. I used Revolut eSIMs. It’s expensive, but most websites worked.
When I called my parents during the trip, they couldn’t believe how modern China is. For most people in the West, China is still a third-world country.
There are 17 cities with 10M+ inhabitants. Most of them were built in the last 30 years.
When I was in Hangzhou 杭州, I felt like I was in an ant house. Very crowded, yet super modern, quiet, and organized.
My wife and I loved the trip so much. We’ll come back next year and get lost in China because there are so many wonderful places to go.
Kai Chen retweeted
Parallelism is one reason we’re moving Lodestar’s compute-heavy work into Zig.
JavaScript workers run in separate V8 isolates. Ordinary objects sent between them are structured-cloned, which can mean copying large inputs and results. Node can avoid some copies with transferable buffers or SharedArrayBuffer, but that requires explicit ownership and synchronization.
Inside lodestar-z, native worker threads share one address space. Once data is in native memory, workers can operate on the same buffers via pointers rather than serializing inputs for each worker.
That gives us tighter control over how work and memory are shared—and less coordination overhead on parallel hot paths.
Oh look! My longtime friend and colleague @Sanjay_Ghemawat is now on Twitter! Give him a follow!
In April, '17, @jsomers of @NewYorker reached out & said he wanted to do a small profile of me & my longtime colleague Sanjay Ghemawat, watch us work for a few hours, maybe dinner, etc.
It came out today. I think it captures our working style really well. newyorker.com/magazine/2018/…
Kai Chen retweeted
👊I've joined the crew at @ClickHouseDB to establish the *ClickHouse Labs* research group. We'll be working side-by-side with ClickHouse devs to cook up new ways to make the system even faster. I'll also be exploring new topics related to PostgreSQL. clickhouse.com/blog/andy-pav…
Kai Chen retweeted
Ello timeline,
Currently looking for work. If you need someone to sculpt computers into moving data around efficiently (databases, vms, backends, compute kernels), I'd love to chat.
Kai Chen retweeted
Hi my frens, see you in Taipei!!!
ETHTaipei 2026 is taking shape.
Tickets are live. The first speakers are up, from Uniswap and CertiK to the Ethereum Foundation and Across Protocol. Partners and community support are showing up too.
There's still room to shape what happens in Taipei.
See what's already live.
ethtaipei.org/
Kai Chen retweeted
Hazard pointer is an interesting data structure that implements a memory management methodology that allows memory reclamation for arbitrary reuse.
This talk from @KernelRecipes 2024 presents a new synchronization API that combines the 2 data structures hazard pointers and reference counters to leverage the benefits of each mechanism. It uses hazard pointers as fast-paths, and falls back to reference counters either explicitly when the reader expects to hold the object for a long time, or when no hazard pointer slots are available. Hazard pointers, being used on the fast paths improve upon the speed and scalability.
Link to the talk -
Hazard pointers with reference counter -
youtu.be/mnQlQXZn5EI?si=c-QW…
Original paper on Hazard pointers -
Hazard Pointers: Safe Memory Reclamation for Lock-Free Objects by Maged M. Michael -
cs.otago.ac.nz/cosc440/readi…
Kai Chen retweeted
Lodestar processes SSZ data throughout every slot. That’s blocks, states, attestations, and more. How do we achieve less runtime machinery and lower memory pressure on this hot path?
In Lodestar-z, Zig’s comptime generates specialized code directly from each SSZ schema. Field types, offsets, and serialization paths are resolved ahead of time.
At runtime, the serializer writes directly into a caller-provided buffer. It still processes each field, but it doesn’t interpret a dynamic type descriptor or allocate per-field intermediate objects.
Do the heavy lifting at compile time so the hot path doesn't have to.
thank you @ethereumfndn
friends, we do have news but a sad one.
zeam grant support from EF has been discontinued.
while we believed that we were doing tremendous work with an excellent team, carrying the PQ implementation effort on our shoulders, unfortunately we have been caught up in this downturn that EF and ethereum core development ecosystem is going through.
without support, we are unable to passionately champion our cause and work, so while we take a step back, we will see how we can continue to involve ourselves in furthering ethereum ecosystem and its goals.
and we thank Ethereum foundation for the support so far. If you would like to support us in anyway, do reach us.
thank you for your attention to the matter 🙂