Sabitlenmiş Tweet
nvnx.gwei
74 posts
nvnx.gwei
@nvnxu
hacker x researcher crafting with zk + ai
Katılım Kasım 2025
450 Takip Edilen5 Takipçiler
Putting Devcon in Mumbai is wild, at least in this market.
I will not be attending, and know many who are doing the same.
Accessibility is an increasingly important trait for frontier tech, and Mumbai was a step backwards.
Marc Zeller@mzeller
I think doing a Devcon in Mumbai during a bear market with an EF running short of momentum will eventually lead to an empty conference and a lot of commentary. Won't attend. No point.
English
Everyone who thinks AI slop will ruin code efficiency/performance is going to be so surprised when everything is absurdly well-optimized John Carmack style machine code.
English
@oskarth @ethstatus this is awesome!
and will be most impactful thing to come out of ef!
English
Yesterday was my last day at the Ethereum Foundation.
Today we are launching EthSystems. We build confidential systems for institutional Ethereum.
I've spent close to a decade building privacy infrastructure in crypto: p2p messaging at @ethstatus, developing Waku protocols at Vac (both now part of @logos_network), mobile proving tooling with @zkmopro, teaching zero-knowledge proofs with my zkintro primer, and advising @ethereumfndn on privacy and access layer strategy. Most of that was aimed at individuals.
The past year at EF's Institutional Privacy Task Force (IPTF) we've been looking at privacy for institutions. On the surface this might seem different, but there are a lot of similarities. There's also a very strong market need for it, and the timing is right.
I've written in the past about the tension and overlap between cypherpunks and institutions. Twitter is not exactly the best medium for nuance, but right now we are at a sensitive point in time: the defaults for the next generation of financial infrastructure are being set, with or without us. I believe we need cypherpunks in the room when that happens.
Excited to start @eth_systems together with my co-founder @motypes and @_rymnc as part of the founding team. See quoted announcement thread for more details.
x.com/eth_systems/st…
EthSystems@eth_systems
Today we're launching EthSystems. We build confidential systems for institutional Ethereum. Institutions want to use Ethereum, but one of the biggest problems is the lack of built-in, modular privacy tools. We were the Ethereum Foundation's Institutional Privacy Task Force (IPTF) for the past year. We had hundreds of conversations with central banks, regulators, tier-one banks, and asset managers, shipping open source work the whole time. Wall Street has found crypto as an asset class, but not yet as commercial infrastructure. Institutions want to run real flows on Ethereum: stablecoins, tokenized assets, settlement. These are businesses with billions of dollars on the line, and no bank will operate in full public view. On a public ledger, confidentiality is the hard part: each party to a transaction should see what it has a right to see, and nothing more. We have a year of proof of work: private bonds, confidential stablecoin transfers, private settlement across chains, the Ethereum Privacy Map, and more. All with protocol specs and security properties, at our website. We've spent a decade working on privacy in crypto. We know there's no silver bullet. Different use cases need different systems, each designed, specified, and hardened properly, and someone has to do that work. That's why EthSystems exists. We're an independent, for-profit company, backed by long-term Ethereum-aligned investors. This is a decade-long transition, and we aren't going anywhere. If you're an institution that wants to build on Ethereum, talk to us. We're hiring: BD in New York, protocol engineers, ops: join@ethsystems.org
English
The quiet part out loud: a lot of institutional and corporate activity has privacy as a necessary dependency, not a nice to have.
Good to see us finally accepting this.
EthSystems@eth_systems
Today we're launching EthSystems. We build confidential systems for institutional Ethereum. Institutions want to use Ethereum, but one of the biggest problems is the lack of built-in, modular privacy tools. We were the Ethereum Foundation's Institutional Privacy Task Force (IPTF) for the past year. We had hundreds of conversations with central banks, regulators, tier-one banks, and asset managers, shipping open source work the whole time. Wall Street has found crypto as an asset class, but not yet as commercial infrastructure. Institutions want to run real flows on Ethereum: stablecoins, tokenized assets, settlement. These are businesses with billions of dollars on the line, and no bank will operate in full public view. On a public ledger, confidentiality is the hard part: each party to a transaction should see what it has a right to see, and nothing more. We have a year of proof of work: private bonds, confidential stablecoin transfers, private settlement across chains, the Ethereum Privacy Map, and more. All with protocol specs and security properties, at our website. We've spent a decade working on privacy in crypto. We know there's no silver bullet. Different use cases need different systems, each designed, specified, and hardened properly, and someone has to do that work. That's why EthSystems exists. We're an independent, for-profit company, backed by long-term Ethereum-aligned investors. This is a decade-long transition, and we aren't going anywhere. If you're an institution that wants to build on Ethereum, talk to us. We're hiring: BD in New York, protocol engineers, ops: join@ethsystems.org
English
@pcaversaccio bold move. this would make tornado pure immutable & unstoppable.
English
i've been one of the very few people consistently reviewing tornado cash dao proposals (and man, this is fucking time-consuming), and i've come to the conclusion that the recent proposals are either malicious or just bullshit quality (and simply adds unnecessary noise).
along the lines i tweeted recently of "no governance is best governance," i've been working on a _terminal_ tornado cash governance proposal that permanently _disables_ tornado cash governance so that no future proposal (malicious or otherwise) can ever be submitted, voted on, or executed again.
at the same time, it adds `unlockAll()` so previously locked torn, including the torn locked specifically to pass this proposal, is never trapped. any torn held directly by the governance contract itself (i.e. dao treasury funds not accounted for in the vault) is intentionally left _inaccessible_ and will remain permanently trapped forever once the governance is sealed.
my ask: please review the proposal and share your feedback (i haven't done any security reviews so far as well as i'm running on little sleep, so any security comments appreciated as well): github.com/pcaversaccio/t…
the proposal is not live yet. i'll only deploy it if i can get sufficient commitments from torn token holders to support and pass it. if executed, torn effectively becomes a meme token.
tornado cash governance must die. long live tornado cash.
github.com/pcaversaccio/t…
English
@zengjiajun_eth does training flower breed classifiers 8 yrs ago count 🫠
github.com/nvnx7/Flower-I…
English
i wonder how many of your guys have tried training a language model.
it seems super fun.
English
@PaulRBerg oh man i token streaming was such a cool use case.
i still remember hacking a private token streaming project inspired by the same.
gg
English
🇮🇳 Announcing the Zcash India Regional Content Bounty! 🎉
Help bring Zcash education to millions by creating educational content in your regional language.
🎁 Prize Pool: $150
🥇 1st - $50
🥈 2nd - $35
🥉 3rd - $25
🎉 8 Quality Participation Rewards - $5 each
📅 Deadline: 24th July 2026
Create content in Hindi, Gujarati, Marathi, Tamil, Telugu, Kannada, Malayalam, Bengali, Punjabi, Rajasthani, or any other Indian regional language. Add Engligh / Hindi subtitles in the video.
Accepted formats:
🎥 Short videos
🎙️ AI-generated videos
🎤 Voice-over explainers
📝 Articles & blogs
📱 Carousels
🎨 Infographics
🤖 AI-generated educational content
Topics include:
• What is Zcash?
• Why financial privacy matters
• Shielded vs Transparent Addresses
• Zcash wallets
• Real-world use cases
• Privacy myths
To participate:
✅ Post your content on X
✅ Tag @ZecHub, @ZcashIND & @Zcash
✅ Use #ZcashINDBounty
✅ Share your post link in our community
Whether you're a creator, designer, educator, or AI enthusiast - we'd love to see what you build.
Let's make Zcash education accessible in every Indian language. 🚀
English
Eager to hear your life advice.
What should I be doing?
English
@colludingnode better to have a filter that only allows randomness from good sources.
English
@nvnxu Privacy on the L1 is something the community wanted forever but it's not so easy to get right
English
Bitcoin and Ethereum can both do one-shot payments with UTXOs, but where they store them is the important nuance.
For Bitcoin, UTXOs are part of the chain's state and nodes must keep the entire UTXO set to verify new transactions. Spending a UTXO looks it up in state. This means UTXOs must be quickly accessible and can only be deleted from state when spent.
The proposed UTXO implementation for Ethereum is different because UTXOs are not kept in state. When created, they are emitted as an event log into history and assigned an index. For spending, the transaction proves that the UTXO exists in history, and to prevent double spending, the chain keeps a single spent bit for each UTXO index, marking is as spent.
Bitcoin optimizes for direct look-ups that need no witness against history. Ethereum optimizes for smaller dynamic state by moving "existence" into history and keeping only the min. information needed to prevent double spending.
The motivation to go with that approach (which is similar to what Peter Todd described as "Delayed TXO Commitments" for Bitcoin) is its impact on partial full nodes.
A partial stateful node that only keeps a minimal subset of the current state doesn't need to store all UTXOs but only the spent-bit structure. This is minimal state for still being able to trustlessly verify/follow the chain.
UTXOs can be implemented in various different designs with different trade-offs.
On Ethereum, they can provide a cheap solution for (private) payments, without necessarily requiring UTXOs to interact with contracts.
It'd be a feature on top, not a replacement of anything.
Toni Wahrstätter ⟠@nero_eth
Native UTXOs on Ethereum. Payments should be one-shot objects, not permanent state. Bitcoin got this right. Ethereum can bring the same idea to payments: prove existence from history, keep only a spent bit in state, and reduce permanent state by ~99.8%. Check out the blog post for details. ethresear.ch/t/native-utxos…
English
@nero_eth i meant both anonymity + confidentiality but starting with stealth addr is good start ig.
now that u mention app layer im grokking with idea if this eip can be made in a certain way to let protocols share a common wider anonymity set rather than their own smaller & weaker set.🤔
English
if with private you mean "unlinkability" then yes, if you mean "untraceability" then no, if not used with some mixer.
The first property is not being able to tell two recipients are the same entity (this is something stealth addresses give you), and the second one is not being able to tell the sender from a set of potential senders. This one is the more difficult part when it comes to privacy. So far, it's mostly done on the app-layer but there are plans to enshrine it.
English
@nvnxu Only the spent_bits are in state and for accumulators or simple the opening roots, we'd use a ring buffer to keep it's state footprint minimal and constant.
English
@leonardoalt @leanprover technical whitepapers now should ship with @leanprover code instead of latex equations
English
I'm so excited for the future of coding:
- We review beautiful formal specs, AI writes @leanprover code and proofs.
- Faster dev, safer software.
- Seamless integration into existing systems, module by module.
All practical today. Feels magical, but somehow it's real.
powdr labs@powdr_labs
powdr autoprecompiles are now formally verified e2e. - Formal spec audited by humans (~500 LoC) - AI writes the optimizer and proves it obeys the spec - Drop-in FFI replacement for the Rust crate Result: on par with the hand-written optimizer, + it's FVed! (link below) 👇
English
can I say something without everyone getting mad?
1X@1x_tech
NEO’s Hands An API to the Physical World
English