Ted Yin (@Tederminant) header
Ted Yin (@Tederminant) profile picture

Founder @lyquorlabs / Co-founder & former Chief Protocol Architect @AvaLabs/ former Research Advisor @chainlinklabs

Silicon Valley, CA
Joined February 2009
Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)Photo by Ted Yin (@Tederminant)
Been heads down for a while. Never left. Just deep in the field, researching and building. Not L1. Not L2. Not app chains. Not bridges. Think of it as a *maker* of oracles and bridges. Customize and run off-chain logic in the contract directly. No new DSL. Just Rust with minimal macros. Solidity-style programming, compatible with ETH APIs and wallets. Zero state overhead. No Kubernetes. No "fake it like Web3 until you make it." Validators are directly incentivized by YOU to serve your contract, with dedication. With composability.
Lyquor /Lye-kwur/ is a decentralized platform for Off-chain Ready smart contracts, hosted with choice. Today’s decentralized apps are stuck between costly on-chain constraints and centralized infrastructure.
5
4
48
4,864
Modularity is great for engineering, but in an early-stage industry it only matters once you grasp the shifting demands and tradeoffs. I once thought it was everything, but it is really about the right abstraction and architecture. Modularity only amplifies the right core.
11
1,110
Yes. But you need a more fundamental redesign of the infra, instead of just making things modular and gluing the systems together. Enter @lyquorlabs.
Whitepaper Reading SBC25 Session 4: Chainless Is "Chainless" the next step for dApps? @brianseong (@0xPolygon ) and @paulgebheim (@SeiNetwork) led a brain storm session on this new architecture. Core Ideas: - It's a new philosophy: enabling verifiable apps without reliance on traditional chain consensus. - The goal is to unlock new application types that are impossible today. - A practical first step could be provable databases (e.g., PostgreSQL). The technical focus shifts to proving system logs and outputs, not just full state. - Many believe future regulations could be a major driver for adoption. - We want to make this abstract concept into production with a demo application. Summary: docs.google.com/document/d/1… Paper: arxiv.org/pdf/2505.22989 Thank you, @brianseong & @paulgebheim, for co-authoring the paper and guiding the discussion! #Chainless #ZK #VerifiableCompute #SBC2025
2
1
13
1,679
L2→L1 “settlement” = same bridge problem, no atomic, predictable composability. Like dodging bullets in The Matrix. “When you’re ready, you won’t have to.” When your system has both selective hosting and composability, you don’t need a bridge. That’s why we built Fate-Constrained Ordering (FCO): docs.lyquor.dev/docs/litepap…
2
7
604
Let’s meet in person there? 🥳
🔥 SGLang x AMD SF Meetup — Aug 22 🔥 We're teaming up with @AMD for a special AI Infra Meetup at SF. Here's what to expect: 1. An instructor-led GPU workshop to get hands-on with LLM infra on AMD GPUs 2. Talks by AMD, xAI, and the SGLang team on open-source roadmap, MoE scaling, and AMD inference optimizations 3. Good food, great people, and real infrastructure conversations If you're building or exploring LLM infra, this is a solid chance to learn, connect, and share! 📅 Friday, August 22, 2025 🕓 4:30–8:30 PM 📍 Shack15, SF Ferry Building 🎟️ Seats are limited — lock yours now: lu.ma/gbfhjvuo #LLM #AIInfra #SGLang #AMD #SanFrancisco #Meetup
1
3
1
45
7,381
Shake it up! 🧪 You’ll use one tool called `shaker` to build and deploy your contracts (aka Lyquids). We'll have our "bartender" Lyquid for the global registry for all contracts.
1
7
1,595
Chains leave the mark. Lyquor does the work. When you stop tinkering around the chain, you start building beyond it. docs.lyquor.dev/litepaper
10
4
47
7,453
Maybe I'm the last one to notice this, but their receipt system seems to just use an incrementing ID in the URL. crypto.token2049.com/asia/re… Change the ID and you can see other orders.
Me trying to be an "early bird" for @token2049: "Pay with crypto" gives me @coinbase to process.🤔 One-time payment → massive USDC allowance approval pop up → funds gone → still no email after 30min 🤯 Web3 payments meet web2 backend: the worst of both worlds.
1
6
1,091
Someone spent $1198 on Jun apparently...
1
2
542
Me trying to be an "early bird" for @token2049: "Pay with crypto" gives me @coinbase to process.🤔 One-time payment → massive USDC allowance approval pop up → funds gone → still no email after 30min 🤯 Web3 payments meet web2 backend: the worst of both worlds.
3
1
9
1,942
It actually takes *more* transactions upfront just to set the allowance. All for a one-time payment that could have been done in one shot, while exposing you to extra risks if you don't even realize how allowances work as a normal user.
1
475
I wish I were as slim as I was in the pre-COVID era.
This quoted post is unavailable.
5
969
Excited to share our recent research FicusDB at IC3. Grateful to collaborate with Ted @Tederminant and Prof. Robbert van Renesse @rvanren on this work.
Replying to @initc3org
Hongbo Zhang from @Cornell spoke about FicusDB— a new authenticated storage system designed to improve blockchain performance without requiring a hard fork. FicusDB supports existing authenticated data structures like the Ethereum Merkle Tree, while minimizing write amplification and copy-on-write overhead The storage system also boosts read performance and reduces the storage footprint by 66% vs. Ethereum's official Geth client
1
1
2
3
1,227
🍺Brewing something groundbreaking but practical for 2025 @lyquorlabs. Tokyo is an awesome place to work from travel, but you need to account for low and tiny tables. Happy new year folks! There's much more fun to come.
1
16
1,080
Happy 4 years of mainnet, @avax fam! Here's to many more milestones together. The final quarter of 2024 is set to be explosive—get ready for what’s coming next! Let’s keep building and making history. 🔺9000🔺
12
56
4
221
11,823
If I remember correctly, @Tederminant had a cool new design for storage layer of blockchain as well!
Replying to @gakonst
Opcode-level parallelism is not sufficient to significantly enhance the performance of the entire system. also requires optimization and coordination at the data storage level. We have an internal implementation by a new fine-grained opcode, on my M2, it can easily handle upwards of millions of ERC-20 transfers. However, once persistence comes into play, the effect is not as pronounced. We are working a nwe solution, spanning from computation and memory caching to persistence. The core idea is to store as much data in memory as possible, minimizing the number of data persistences and opening additional threads dedicated to handling tasks at the data engine level, such as hashing and indexing. Previously, we were developing based on Geth, and the new version will be based on Reth.
1
2
5
4,420
Well...not so cool @_patrickogrady. After months of single-person effort of me proposing the architecture and finishing the first MVP alone, you guys mentioned me in 20th tweet in the thread, removed me from the author list in the repo, and redrew the pictures that I drew.
Replying to @_patrickogrady
20/ Thanks to @Tederminant, @rkuris, @m_s_p_h_x, @_richardpringle, and Dan Sover for their contributions to this project over the past few years. We'll talk about this project at length at the next Developer Community Call! See you there!
22
7
4
110
45,005
and honestly speaking the new graphics for the architecture looks worse than the original one lol.
6
25
4,927