iamlogtun.eth

@puffer_finance Finally a restaking solution built for enterprise needs. Well done Puffer! 🔥🐡

👔 Institutional Staking #2 - How Institutions Access ETH Yield Without Giving Up Control Most ETH staking products give institutions pooled exposure. Puffer Institutional gives them a dedicated vault. This matters because institutions can separate their position, define permissions, and control how ETH moves through staking and restaking. When institutions deposit ETH, they receive ITokens representing their specific staking and restaking position. Behind that position, the Institutional Vault manages validator operations, coordinates with the Restaking Operator for @eigencloud restaking, handles reward accounting and withdrawals, while AccessManager defines who can perform each key action. The architecture also separates staking rewards from restaking rewards, giving institutions clearer accounting across validator yield and AVS participation. That is how institutions access ETH yield without giving up control: dedicated vault infrastructure, ITokens representing their position, configurable permissions, direct validator operations, and structured restaking inside one controlled system. ETH yield, structured for institutional control. 🐡

@StecoinZhang @puffer_unifi Puff puff❤️

@iamlogtun @puffer_unifi 🐡🐡

Ethereum scaled with rollups but fragmentation came with it. What if appchains could scale independently, while still feeling like one unified chain? No bridging, no siloed liquidity nor fragmented UX @puffer_unifi enables synchronously composable based appchains powered by @ethereum itself. @puffer_finance @puffer_unifi @puffer_preconf @AmirOnchain @StecoinZhang @art3chris

@BaseHubHB @virtuals_io @AskVenice @clanker_world @flock_io @shadowcleague @axisrobotics @roarin_ai @axisrobotics 🔥

Which AI on Base do you like best?

@axisrobotics Always Axis

Which robotics project on Base do you like best?

@iamlummyjay @puffer_unifi Thanks fam🫡

@iamlogtun @puffer_unifi Well detailed. Enjoyed listening bro.

@ChanduS279 gAxis fan

Most people still analyze robotics from the hardware layer. That’s outdated. The real bottleneck is: trajectory data simulation scale behavioral diversity policy iteration speed Physical AI is increasingly becoming a data infrastructure problem. That’s why browser-based robotic training systems are architecturally interesting. @axisrobotics

@TeddyMoney001 @puffer_unifi gmgm fam

@iamlogtun @puffer_unifi Morning boss

@Larrizle_no1 @puffer_unifi thanks fam

@iamlogtun @puffer_unifi Nice content chief

@aandooor @puffer_unifi puff puff🐡❤️

@iamlogtun @puffer_unifi unified app chain yes please 🐡✨

@shihyumo @puffer_unifi thanks fam🐡❤️

@iamlogtun @puffer_unifi Puff puff🐡 this is awesome!

@pufferintern @puffer_unifi Puff puff🐡❤️

@iamlogtun @puffer_unifi great video, fren

@StecoinZhang Solid breakdown here 🐡🔥

Many people think synchronous composability just means “fast bridging.” It does not. It means solving three separate infrastructure problems simultaneously for based appchains. 1. L2 → L1: withdrawing from L2 to L1 without waiting long time. 2. L1 → L2: accessing L1 liquidity from L2 without trusting a third-party bridge. 3. L2 → L2: coordinating atomic execution across separate based appchains. Starting with L2 → L1 — the capital efficiency problem. The problem: rollup withdrawals require verification that the rollup state is valid before Ethereum releases funds. Optimistic rollups enforce a 7-day challenge window. Even ZK rollups can take minutes or even hours. Capital becomes trapped between layers. Institutions cannot operate efficiently. DeFi liquidity fragments. The @puffer_unifi solution: proof-generated bridging. TEE provers — Intel TDX and AMD SEV — independently re-execute the rollup state transition. Both must agree. The L1 rollup contract verifies the proof on-chain. Funds are released atomically. No long time wait. No third-party attestation. No off-chain verifier to compromise. TEE is live today. ZK proving is next. As ZK matures, it joins the prover multisig alongside TEE, adding a second independent trust assumption. The fast withdrawal from L2 to L1 is not released because time passed, but it is released because the rollup state was verified. Next up: L1 → L2 and L2 → L2. 🐡

@bashcaliz4u @axisrobotics Nice thread fam

Why Data Is the Backbone of Robotics 🤖 The AI boom was built on internet-scale data. The robotics revolution will be built on something far harder to obtain: Physical interaction data. And this is exactly where projects like @axisrobotics are focusing their vision. 🧵 👇

@Sam_wolf1122 @axisrobotics Where do you see the funding details bro?😎

Why is everyone talking about @axisrobotics? 1️⃣ The Tech: A 6-axis control stack for total deterministic movement. 2️⃣ The Lead: Dr. G, a veteran in industrial kinematic control. 3️⃣ The Backing: $5M seed round led by Galaxy Here is a professional breakdown 🧵👇

@Sam_wolf1122 @axisrobotics Nice content bro

@OXEMMYWEB3 gAxis fam

𝗪𝗵𝘆 𝗗𝗮𝘁𝗮 𝗶𝘀 𝘁𝗵𝗲 𝗕𝗮𝗰𝗸𝗯𝗼𝗻𝗲 𝗼𝗳 𝗥𝗼𝗯𝗼𝘁𝗶𝗰𝘀 Everyone understands why ChatGPT needed the internet. Billions of web pages, articles, books with all of it scraped, cleaned, and fed into a model until it could think in language. Robots need the same thing. But there's a catch: The internet doesn't teach a robot how to move. LLMs run on text. Robots run on physical interaction data with how to pick up an object, navigate a space, respond to resistance, recover from failure. That data doesn't exist on Google. It has to be generated. And that's exactly the gap Axis is filling with one contributor task at a time. The data problem for robotics is the biggest bottleneck in physical AI right now. Whoever solves it, wins. (@axisrobotics)

@puffer_unifi Solid infrastructure 🐡🚀

The lookahead is what makes based sequencing credible. Gateways know their slot windows in advance. State syncs before transitions. Preconf issuance remains continuous across every slot. For UniFi appchains, this means L1 and L2 share not just a sequencing root, but a predictable, coordinated execution environment where users always know who is sequencing and when. Deterministic by design. Not a soft promise. 🐡

@puffer_preconf Solid infrastructure play by the Puffer team. 🐡🚀

📖 Glossary: Lookahead Mechanism A schedule that maps upcoming Ethereum L1 slots to their assigned proposer, who delegates sequencing rights to a gateway, revealed approximately one epoch (~6.4 minutes) in advance. The lookahead mechanism gives rollups predictability over who provides preconfirmations and when. Before each slot transition, the outgoing gateway syncs state with the incoming one, ensuring continuous preconf issuance without gaps. Puffer Preconf translates that structure into fast, coordinated execution across rollups. 🐡

@iamlummyjay @axisrobotics Nice effort, ser

while touching grass, let’s learn a little about AXIS everyone talks about AI robots like the hard part is the model. Is it really? the bottleneck is data. LLMs had the internet. physical AI has almost nothing. that’s the gap @axisrobotics is trying to solve @iamlogtun