Web3 explorer & content creator !!π© Certified Creator @starlisting collaborations & sponsored !! TG π https://teapaw.org/t.co/TE44wJGaUn
South Asia
Joined March 2025
- Tweets22K
- Following7.4K
- Followers9.1K
- Likes47.1K
ππ¬π₯π¨π¦ retweeted
Every creator I know has a clip that did well in one language. Dub & Sync is going to give a lot of those a second life
π€ Paid partnership
ππ¬π₯π¨π¦ retweeted
Ngl, I expected zkTLS to be one of those things that sounds sick on CT and is painful to use.
I've been on Primus since , I've done the tasks, and it wasn't that. Then I went down the FHE rabbit hole and it got more interesting.
Start with zkTLS, the part I actually used.
I didn't have to ask any platform for anything. No integration, no partnership. zkTLS works with the websites, APIs and exchanges that already exist, and most privacy tech dies at exactly that step.
So I ran my own flow. The extension generated the proof from my own browser session. I never shared a login and never sent a screenshot. The app learned one fact, and the rest of my data stayed with me.
Every other time I've had to "prove" something in crypto, it meant screenshots anyone can edit or access I shouldn't hand over. A proof is neither fakeable nor risky.
For builders, there are two modes. MPC gives stronger protection during TLS session setup. Proxy is faster, with different trust assumptions. You pick the trade-off based on what you're proving.
That's the verify half. Here's the compute half.
zkTLS proves a fact is real. FHE lets you use it while it stays encrypted. It never took off because it was too slow, and Primus published a technical deep dive on how they fixed that:
β A full confidential transfer on encrypted 64-bit balances finishes in about 90 ms
β One bootstrapping takes about 5 ms
β 210+ confidential transfers per second on 8 consumer RTX 5090 GPUs, scaling linearly
The problem: classic FHE processes one tiny gate at a time, and every gate needs a noise refresh called bootstrapping. One 64-bit operation becomes thousands of them.
The fix: pack more work into each bootstrapping, then use AutoHoG, a compiler that takes standard Verilog and collapses clusters of gates into single custom gates, cutting latency by up to 5.7x. The GPU engine then runs one uniform operation, which is exactly what GPUs are good at.
One detail I thought was underrated: they use exact integer arithmetic instead of floating-point, so results are byte-identical across machines. For decentralized consensus, that matters a lot.
And it runs on consumer GPUs you can buy off the shelf.
Privacy usually loses because it costs too much. When a confidential transfer takes about 90 milliseconds, privacy stops being a feature you have to justify and becomes the default.
Put the two halves together: zkTLS proves facts from the platforms people already use, and FHE computes on the result without decrypting it. Both are getting fast enough to feel like normal software. With 1M+ users and 10+ paid enterprise customers, this isn't staying in the lab.
Proof without exposure isn't a slogan once you've used it yourself.
What would it take for you to use privacy by default: speed, cost, or not having to think about it?
@primus_labs
ππ¬π₯π¨π¦ retweeted
An AI agent is only as useful as the tasks it can actually perform.
Giving an AI the ability to understand instructions is one thing. Giving it a way to interact with blockchain applications while respecting the owner's rules is another.
This is an interesting problem @meetmotes is exploring.
Motes brings together AI agents, NFTs and smart accounts to create a system where agents can have their own identities and interact with onchain applications.
Why is this important?
1. Practical utility: AI agents could move beyond answering questions to handling assigned activities.
2. User control: Owners can define permissions and limits for their agents.
3. Onchain access: Smart accounts provide a way for agents to interact with blockchain applications.
4. New possibilities: It opens up more ways to combine AI automation with DeFi.
The real opportunity isn't simply making AI smarter. It's making AI capable of doing useful work without removing human control.
That's the direction I'm watching with Motes.
Explore the campaign:
hazels.io/studio/pulse/motesβ¦
ππ¬π₯π¨π¦ retweeted
Connecting a phone to @Acurast is only the beginning. Keeping it online is what gives it more time to contribute.
The Nucleus campaign considers active device duration when calculating performance multipliers. It also counts compute from people you refer, so helping them keep their Processors online can contribute to your overall standing.
I find this more interesting than simply connecting as many phones as possible on the final day.
A device that stays active has more opportunity to contribute than one that repeatedly disconnects.
Consistency matters in decentralized compute, too.
ππ¬π₯π¨π¦ retweeted
230 users in Hypesium's first week on @Enter_Elysium.
They trade tokens from 19 launchpads and bridge across 7 chains without leaving a Telegram chat.
Paste a token's contract, tap Buy.
t.me/hypesium_bot
ππ¬π₯π¨π¦ retweeted
πGrading is fully caught up.
Over the past 2 days, posts waited longer than usual for their AI grade while we upgraded the platform. That backlog is cleared: every post and reply is now graded and the boards are up to date.
If your STARS moved, that is why.
No post was missed.
ππ¬π₯π¨π¦ retweeted
Episode 3 Launchpad Compare
vs. StonkFun - Stonk
Both pair coins with stocks. Both pay holders.
StonkFun: Solana only, bonding curve first. Buys pay 1%, plus up to 3% tax.
Rebate Pad: live Uniswap v4 pool, pay from six chains. Buys add bonus tokens, 2 to 9%.
Their site, our docs, same table. π
ππ¬π₯π¨π¦ retweeted
GM FamilyβοΈ
The most interesting thing about @vangrid_io isn't just that people can capture the physical world with their phones.
It's that someone can actually request a specific place, fund the task, get a real person to capture it, review the result, and receive a reconstructed 3D model.
That turns spatial data from something collected in advance into something that can be requested when a real need appears.
And the marketplace is showing this isn't limited to one type of environment.
Recent requests include a multi-storey car park in Glasgow for $1,500, a station concourse scan for $1,200, and a fleet depot wash bay that was accepted for $1,200.
The latest board shows 1,116 bounties and more than $744K in recorded settlements.
Those numbers matter, but the bigger signal is the variety of physical spaces people are willing to pay to have captured.
For robotics, logistics, mapping and Physical AI, the world is constantly changing. A useful data layer needs a way to keep up with those changes, not just rely on yesterday's datasets.
That's the idea I find worth following.
A distributed network of people capturing real environments, with requests, payments and proof connected through an open marketplace.
Good work from the Vangrid team and everyone contributing to the campaign.
This may be the end of this campaign, but the need for real-world ground truth is only getting started.
π€ Paid partnership
ππ¬π₯π¨π¦ retweeted
Most people think of cloud computing as something that happens somewhere inside massive data centres.
Rows of servers, expensive hardware, enormous electricity consumption and infrastructure controlled by a handful of major companies.
But what if computing resources could come from somewhere much closer to home?
Your smartphone.
Thatβs the interesting idea behind Acurast.
Acurast is building a decentralized computing network that allows supported smartphones to contribute their processing resources to a shared infrastructure.
These devices become what the protocol calls Processors.
Instead of computing being concentrated entirely in traditional data centres, Acurast coordinates independent devices to execute workloads for developers and applications.
And this goes beyond simply connecting phones to a network.
The objective is to make these devices useful for actual computing tasks, including backend services, APIs, AI inference and confidential applications.
Think about the implications.
There are billions of smartphones around the world. A significant portion of that hardware remains underutilized for much of the day.
@Acurast is exploring how some of those resources can become productive infrastructure.
Of course, smartphones wonβt replace every high-performance server or GPU cluster. Thatβs not the interesting part.
The interesting part is creating another way to access computing resources.
A different infrastructure model, powered by devices people already own.
Thatβs where Acurast comes in.
Website: acurast.com/
Powered by: @NucleusCodes
What if the next generation of cloud computing doesnβt depend entirely on massive data centres, but on the smartphones already sitting in peopleβs pockets?
Thatβs the idea behind @Acurast
Acurast is building a decentralized computing network that transforms everyday smartphones into distributed computing infrastructure.
Hereβs why that matters.
Today, much of the worldβs computing infrastructure depends on centralized cloud providers. Developers rely on a relatively small number of companies to host applications, process information and provide the resources their products need.
Acurast is exploring a different approach.
Instead of concentrating computing power in traditional data centres, it allows individuals to contribute supported smartphones to a shared network where developers can deploy and execute computing workloads.
These connected devices are called Processors.
But Acurast isnβt simply turning smartphones into mining devices. The objective is to make their computing resources useful for real applications.
Think AI inference, backend services, APIs, blockchain automation and confidential data processing.
Thereβs also an interesting security component.
Acurast incorporates Trusted Execution Environments (TEEs) and hardware-backed attestation to help protect workloads and verify that computing tasks are running on genuine, appropriately secured devices.
This matters because decentralized computing isnβt just about distributing hardware. Developers also need confidence in how their applications execute and how sensitive information is handled.
The ecosystem has its own native token, $ACU, which supports network fees, staking, provider incentives and governance.
For individuals, there are opportunities to participate as compute providers by connecting supported smartphones. Others can explore the ecosystem through its developer tools and community initiatives.
One of those initiatives is Cloud Rebellion, where participants engage with the ecosystem, complete activities and accumulate MIST through eligible campaign tasks.
What makes Acurast interesting to me is the infrastructure problem itβs trying to solve.
We already have billions of smartphones distributed across the world. Acurast is exploring how some of that existing hardware can become part of a globally accessible computing network.
Of course, the real test is whether this distributed supply can deliver reliable computing services that developers consistently want to pay for.
Thatβs the part worth watching.
Cloud computing is evolving, and Acurast is taking a rather different route.
Explore the ecosystem:
πΉ Website: acurast.com/
πΉ X: @Acurast
πΉ Cloud Rebellion: rebellion-backend.acurast.coβ¦
Perhaps the next cloud wonβt belong to a handful of data-centre operators. Perhaps it will be powered by a network of devices distributed across the world.
ππ¬π₯π¨π¦ retweeted
Instagram likes have consequencesβ¦ and apparently, they come with a diamond price tag! ππ
Created with @Hailuo_AI using MiniMax. π¬
π€ Paid partnership
ππ¬π₯π¨π¦ retweeted
An AI trading agent can execute strategies, but how does it prove that it deserves access to capital?
This is the problem @agenticscredit is working to solve with its Agentic Credit Score (ACS).
Instead of judging an agent by a single PnL screenshot or the amount of capital in a wallet, ACS evaluates its actual trading history through six factors:
β’ Profitability
β’ Drawdown
β’ Consistency
β’ Longevity
β’ Win Rate
β’ Sharpe Ratio
The score ranges from 300 to 850 and updates as trades close. Both paper trading and real on-chain activity contribute, but real-money performance carries the greatest weight. This means a strong paper record cannot simply hide poor live trading performance.
The interesting part is what happens at the 580 threshold.
A score of 580 or higher qualifies a wallet for the constrained credit path and funding waitlist. Below that threshold, traders can build their track record through paper trading under the platform's risk engine.
Importantly, reaching 580 is a qualification milestone, not a guarantee of immediate funding.
I think this is an important direction for agentic finance. AI agents need more than execution capabilities. They also need measurable performance, risk discipline, and a credit history that others can evaluate.
Would you trust a trading agent based on its highest PnL, or its long-term consistency and drawdown record?
Explore your ACS: agentics.credit
More on @agenticscredit.
π€ Paid partnership
ππ¬π₯π¨π¦ retweeted
most smartphones spend a lot of time doing nothing.
there might be an old android sitting in a drawer, switched off for months. instead of letting it collect dust, you could turn it into a dedicated Processor through @Acurast
with Processor Core, a compatible spare android can contribute computing power to the network. it requires a factory reset, so it's better suited for a device you no longer need for everyday use.
if it's your main phone, Processor Lite offers another approach. compatible android and iphone users can contribute during idle periods, like overnight charging, without giving up their apps or personal data.
i'd choose Core for a spare phone and Lite for my daily device. different setups, same idea: put unused computing resources to work.
ππ¬π₯π¨π¦ retweeted
Finally starting my Kaito journey! π
Day 1 and I've got 20 Aura with 40 Smart Followers.
It's a small start, but every big journey starts somewhere. I'm excited to explore Kaito, share my thoughts, stay active, and see how far I can go.
Not rushing anything, just learning, contributing, and enjoying the process.
Let's see what the journey brings!
Tell me yours π«΅
Explore here - s.kaito.ai/2EPNJD
ππ¬π₯π¨π¦ retweeted
What actually happens in the few milliseconds after a liquidity pool is created?
@Hypesium has an interesting answer in its execution architecture.
The project describes a pipeline designed to minimize work at the moment of execution.
Before the trigger:
β Route construction.
β Calldata preparation.
β Gas configuration.
β Transaction signing preparation.
After the pool creation log:
β Nonce validation.
β Final signing step.
β Submission across four lanes.
Those lanes consist of:
βΈ Two WebSocket connections.
βΈ Two keep-alive HTTP connections.
The documented hot path targets under 5ms from the pool event to submission.
What I find interesting is the engineering principle behind this.
Instead of making every decision at the last moment, the system prepares predictable operations beforehand.
It's a good example of why execution architecture matters just as much as raw network speed.
The next question is how consistently those four lanes perform under real network conditions.
ππ¬π₯π¨π¦ retweeted
I found something interesting in @Acurast's recent AI test.
They managed to run Qwen3.5-9B, a 9 billion parameter AI model, on Android phones with 16GB RAM.
On the fastest phone, it generated around 9.3 tokens per second. They also tested running the model across more than 100 phones.
But there's one detail worth mentioning. Processing a long prompt of 7,000 tokens still took over 3 minutes on the fastest device.
So I don't think phones are ready to replace powerful AI servers for every task.
What makes more sense to me is using many phones together for smaller AI jobs that can run in the background.
That's what I find interesting about Acurast. It's not just about connecting thousands of phones, but finding useful things those phones can actually do.
ππ¬π₯π¨π¦ retweeted
Good morning CT
Been watching how @Hypesium is trying to attract traders into the Elysium ecosystem.
They just announced $1,850 in weekly rewards for their top 3 referrers, paid in $HYPE.
But here's the part worth discussing.
Your ranking depends on the fees generated by the people you invite.
Someone with 10 active traders could outperform someone with 100 inactive referrals.
For a trading bot that only went live a few days ago, this is a pretty interesting way to build an initial user base.
The real test will come when traders decide whether the execution speed, automated exits and overall experience are good enough to keep using it.
That's what will determine whether Hypesium can retain users beyond the reward campaign.
@starlisting
$1,850 every week for the top 3 referrers on Hypesium.
π₯1st: $925
π₯2nd: $555
π₯3rd: $370
Paid in $HYPE. You are ranked by what your referrals earn you: up to 45% of their trading fees on Elysium or HyperEVM.
Round 1 ends Friday Nov 20, 20:00 UTC.
Get your link: /referral in t.me/hypesium_bot
ππ¬π₯π¨π¦ retweeted
opened the vault page on @rntprotocol and kinda paused at the withdrawal line...
you deposit usdc, and it goes into a mix of real estate strategies. that's rwa... real stuff like buildings, but onchain. it's on ember, run by tulipa capital.
but you don't get your money back the same day. withdrawals are monthly. the page says the next run is 1 dec, 05:00, and they aim to send it back in about 45 days, or up to 60 if it's busy.
it still shows 15.23% apy, last updated nov 2, so that number can move. there's also a 20% fee on the profit.
has anyone actually waited through a withdrawal on @rntprotocol yet, or just watched that date?
ππ¬π₯π¨π¦ retweeted
Happy Weekend X family.
What if the next viral coin came from something nobody thought twice about?
That's the creative opportunity @onlystreetsfun is bringing to Solana.
A funny shop sign, an unusual street scene, a random object or an unexpected moment could become the starting point for a new coin.
What makes OnlyStreets different is its creation rule: every coin must begin with a fresh photo captured through its in-app camera. You cannot simply upload an old meme or pick an image from your gallery.
I find this interesting because it makes discovery part of the launch process. You actually have to go out, look around and find something worth turning into internet culture.
With @rarelistio I'm looking forward to seeing what creators come up with.
My question is: What is the most random thing in your neighbourhood that you think could become a memecoin?
π€ Paid partnership
