Web3; Educationalist||Explanatorist||Storyteller|| content writer.
Joined June 2015
- Tweets3.8K
- Following961
- Followers1K
- Likes7.3K
Pinned Tweet
A stock on a blockchain is still just a stock.
Tokenizing solved who can buy it. It didn't build anything around it, and that gap is why I joined the @2FactorFinance Points Program.
Two terms first. Capital structure is how risk and cost get split between holders. A tranche is a slice of an asset's risk with its own rules.
→ Access is solved. As of September 19, Coinbase's tokenized stocks had crossed $1 billion in decentralized exchange volume over 30 days, per 2Factor's announcement.
→ Structure isn't. Who carries which risk, and who pays for the leverage?
→ 2Factor splits an asset's volatility into two perpetual tranches: a senior side with deep downside protection, and a leveraged junior side with no liquidation risk and no external hedging counterparty. The junior side pays the senior side a premium for the leverage.
The mismatch shows up in sizing. A fixed 2x or 3x ignores volatility. At Bitcoin's volatility the range where leverage helps compounding closes early, so the junior targets about 1.33x for cbBTC, and calmer assets get more room. Cost is the other half: conventional leverage is priced against the cost of shorts, and 2Factor prices it against the cost of stable capital.
2Factor is on Base and works together with Coinbase to bring Kelly-optimized leverage onchain, a good use case for Coinbase's tokenized stocks ecosystem. It's live on Base as a capped-liquidity beta, Bitcoin first, with tokenized securities a later phase. The exposure still carries risk.
I joined the Points Program. Season 1 rewards completed actions: social activities, education and referrals. Marks come only from verified actions, so buying, depositing or holding earns none, and Marks have no cash value. The top 10 on the leaderboard split 1 BTC, paid in cbBTC, subject to eligibility checks. Join with my link if you want in. I may earn Marks if you do. 18+, not open in the US or other restricted regions, terms apply.
points.2factor.finance/r/289…
My view: tokenized assets will be judged by the structure around them, not by how many people can buy them. If tokenized stocks need more than access, what's the first piece of structure you'd add?
Most people think owning something means you're free of everyone else.
Most of what we call owning is a promise someone else has to keep. @0xhazels has a poem line from Rin, tied to Money, that says it better than a finance textbook: "Learn to owe your future to no one."
Here's the plain version.
→ An IOU is a written promise to pay. A bank balance is one: a number in the bank's database, backed by its promise to hand you the money.
→ Counterparty risk is the chance that the other side doesn't keep its promise.
→ An asset recorded on a public blockchain and held in your own wallet depends on code and keys instead. The rules are written down where anyone can read them.
Now follow the chain of promises.
Every extra party between you and the thing you own is one more promise that can break.
An app may hold it through an exchange, the exchange through a custodian, and each one has its own ways to fail.
When one fails, you don't get to act. You wait in line.
Hazels' contract shows as verified on Etherscan, which means its source code is public. That's a rule you can read instead of a promise you have to trust.
This isn't a claim that self-custody is safe. You carry other risks, like lost keys and bugs, and no one can reset them for you. It's a claim about what you depend on.
So the contrast:
a promise you have to trust,
or a rule you can read.
My bet: the next wave of users will ask who they depend on before they ask what they own. How many promises sit between you and the thing you think you own?
Most people think value works like money: pass it on and you have less.
Not everything does. @0xhazels has a line from Yuki: "Value grows as it changes hands."
I don't mean price. A coin runs out when you give it away. A skill doesn't: I teach you, I still have it, and now you do too. Hoard what you know and it leaves with whoever leaves.
My bet: the onchain communities that last will reward teaching. What's one thing you learned faster by explaining it to someone else?
On an early-stage campaign, most people haven’t heard of the product yet. @rarelistio says it’s launching Agent Dojo at an early stage.
So the job isn’t promoting. It’s translating.
→ First, one sentence: what is it? Agent Dojo: train AI agents and send them into battle.
→ Then one specific: what would I do with it?
Hype tells people to look. Translation tells them what they’re looking at.
My prediction: early-stage campaigns reward the clearest explainers.
Posting this on @rarelistio.
@rarelistio just wrapped the referrals campaign, watching out for the rest of the campaigns
@agentdojo, @projectspherexy ,@fomopoly_rh and the @rarelistio season 1 campaign
You can still hop in now:rarelist.io/r/onchainscribe_
Onchain Scribe retweeted
Happy Thursday Friends..
Many of you have DM’d me asking how I reached the Top 09 in the @rarelistio campaign. 👀
With 9,184 creators competing for the Season 1 $50K reward pool, it looks like a tough race. In practice, the approach is simple:
✅ Complete every task the campaign gives you.
✅ Submit one or two high-quality pieces of content.
✅ Stay consistent.
That’s how I climbed the leaderboard. Rarelist rewards quality over quantity, so a few thoughtful posts beat a lot of rushed ones.
Keep it simple and keep building. 🚀
Join now: rarelist.io/r/atiqur2904
Good Morning Friends
Rarelist is setting the pace 🔥
Campaign ended, rewards already delivered. They’re among the fastest when it comes to getting rewards into users’ wallets.
Execution speaks louder than promises, and more opportunities are already on the way. Bullish on what’s next! 🚀 @rarelistio
Join now, its still early:
rarelist.io/r/atiqur2904
Onchain Scribe retweeted
Congrats to @The_DTCC on the launch of its Tokenization Service. We’re proud to be among the firms collaborating with DTCC.
@the_matter_labs is providing privacy features with Prividium.
DTCC provides custody and asset servicing for more than $114 trillion in securities.
Onchain Scribe retweeted
Good Morning Friends
Rarelist is setting the pace 🔥
Campaign ended, rewards already delivered. They’re among the fastest when it comes to getting rewards into users’ wallets.
Execution speaks louder than promises, and more opportunities are already on the way. Bullish on what’s next! 🚀 @rarelistio
Join now, its still early:
rarelist.io/r/atiqur2904
Our short bonus campaign is officially over!
rarelist.io/projects/rarelis…
Thanks to everyone who took part.
Check your ETH wallets, rewards are already out!
More opportunities are already on the way.
Onchain Scribe retweeted
@rarelistio just wrapped the referrals campaign, watching out for the rest of the campaigns
@agentdojo, @projectspherexy ,@fomopoly_rh and the @rarelistio season 1 campaign
You can still hop in now:rarelist.io/r/onchainscribe_
Our short bonus campaign is officially over!
rarelist.io/projects/rarelis…
Thanks to everyone who took part.
Check your ETH wallets, rewards are already out!
More opportunities are already on the way.
Onchain Scribe retweeted
GM CT
Keep an eye on @projectspherexy for the release of the first star.
The first star is already waiting.
$SPHERE arrives on Solana on November 3.
Until then, there is no official token.
Onchain Scribe retweeted
Our short bonus campaign is officially over!
rarelist.io/projects/rarelis…
Thanks to everyone who took part.
Check your ETH wallets, rewards are already out!
More opportunities are already on the way.
GM CT
Keep an eye on @projectspherexy for the release of the first star.
The first star is already waiting.
$SPHERE arrives on Solana on November 3.
Until then, there is no official token.
Onchain Scribe retweeted
The first star is already waiting.
$SPHERE arrives on Solana on November 3.
Until then, there is no official token.
Onchain Scribe retweeted
We assume the first lesson in any discipline is how to win.
In jujutsu, it's how to fall. @0xhazels builds its story on that: the first lesson is how to rise after being thrown. My read: beginners who can't fall get hurt, then stop coming back.
Most brands sell the win. Hazels starts with the throw, and its eighth rise is getting up together.
My prediction: the communities that last teach members how to get back up. What last threw you, and what got you back on your feet?
Random thoughts are good wallet security means never making a mistake.
I think it means making sure one mistake can't reach everything. @0xhazels has a poem line that fits: "Falling is not defeat. Staying down is."
Its site says the first lesson at Azura Jujutsu is how to rise after being thrown. Not how to avoid every throw.
Here's the plain version.
An approval is a permission you sign that lets another address move your assets. It stays active until you revoke it. Blast radius is how much one mistake can reach.
→ Everyone makes a mistake eventually: one wrong signature, one fake site.
→ With one wallet for everything, the wallet you tried a new site with is the same wallet that holds what matters.
→ With separate wallets, a small one for trying things and another that holds what you can't afford to lose and rarely signs anything new, a bad approval only reaches the small one.
Now follow the single wallet.
You connect it to something unfamiliar.
You sign an approval you didn't fully read.
The permission stays active.
And everything that wallet holds is now inside the blast radius.
Hazels' own addresses page shows the same idea at the project level: separate wallets for separate jobs, including owner, payout, royalties and a vault.
This isn't advice for any one person's setup, only a habit worth learning early.
So the contrast:
one wallet, one point of failure,
or separate wallets, so a fall is something you can get up from.
My take: the next wave of onchain users will learn wallet separation in week one, the way people learn passwords. What's one habit that has saved you from a bad signature?
Onchain Scribe retweeted
My thought was that to pay for anything online, you first need an account.
For software agents, an account is the wrong shape. The web already has a status code for this, 402 Payment Required, and @injective supports a protocol built on it called x402. Here's how an agent buys something without signing up 👇
→ The ask
The agent requests an API endpoint. Instead of data, the server replies with a 402 and a price.
→ The proof
The agent signs a USDC payment and retries the request with that proof attached. The server checks it and returns the data. No API key, no subscription, no registration.
→ The speed
On Injective EVM, payments settle in a single block, about 650 milliseconds, per Injective's docs. Their demo fetches market data gated behind a 0.01 USDC payment.
Now the stakes.
An agent that needs five data sources can't stop to register with five vendors. Each signup is a human step. And a subscription makes no sense for software that might call an endpoint once. Pay-per-request fixes that, but it only works if the fee is smaller than the price, which is why settlement cost matters. It also means an agent that can pay can overspend, so the whitepaper's mention of policy-bounded signing is part of the design, not an extra.
The contrast is what stuck with me:
Subscriptions are built for people who come back.
Pay-per-request is built for software that shows up once, pays and leaves.
My prediction: the agent API economy will be priced per call, and the chains that win will be the ones where settling a call costs less than the call.
Onchain Scribe retweeted
Be cautious with Starlist. There are concerns that it may be copying the concept and work of Rarelist (@rarelistio).
Before engaging, connecting your wallet, or sharing personal information, DYOR and verify the project's legitimacy. Don't risk your assets based on hype.
Stay vigilant. Protect your wallet. Verify before you trust. 🛡️
Onchain Scribe retweeted
I used to think token issuance was a schedule: a number set once and left alone.
On @injective it works more like a thermostat. The October 2026 research paper describes a dynamic issuance controller, and once you see it as a feedback loop, the design makes sense 👇
→ The target
The network aims for 60% of INJ to be bonded, meaning staked to secure the chain. Governance proposal 617, passed in January 2026, kept that 60% goal.
→ The correction
When bonded participation falls below the target, the controller raises the supply rate to pull more stake in. At the target, the rate stays unchanged. Above it, the rate eases.
→ The guardrails
Proposal 617 also set annual issuance bounds of 2.2% and 4.4%. The controller can move, but only inside a range governance has chosen.
Now the stakes.
Bonded INJ is what gives validators their voting power, and finality depends on more than two-thirds of that power committing to a block. So participation isn't only an incentive question. It's a security question. A fixed schedule can't react if staking drops. A controller can.
The contrast is what stuck with me:
Issuance pays people to secure the chain.
The Community BuyBack burns are a separate flow, turning ecosystem value into removed supply.
To be fair, the controller only moves the supply rate. It promises no outcome for holders, and staking rewards are financed by that issuance. This is not financial advice.
My prediction: as tokenized assets raise the stakes on settlement security, chains will compete on how well their security budget adjusts itself, not on how big it is.
Onchain Scribe retweeted
We assume DeFi asks for more collateral than the loan because the code is cautious.
I think it's because every wallet looks identical. Collateral is how a lender protects itself from not knowing who it's lending to. @primus_labs, Brevis and BNB Chain built ZKredit to change what a lender can know. Here's how 👇
→ The proof
Primus's zkTLS verifies that data from a real platform, like a centralized exchange, is genuine. A browser extension generates the proof on the user's device, and only the proof goes onchain.
→ The registry
Verified data points sit in a neutral registry on BNB Chain. A lender can check that an account has been active for three years or clears a balance threshold, without seeing a single number.
→ The guardrail
Each Web2 account can be bound to only one wallet globally, with a cooldown before rebinding, so one good record can't back a swarm of wallets.
Now the stakes.
If a lender can't tell a careful borrower from a stranger, everyone gets the same terms: lock up capital first. That capital sits idle, and a strong real-world record counts for nothing. The obvious fix, sending the raw history onchain, is one almost nobody will accept.
The contrast is what stuck with me:
Collateral protects the lender from not knowing.
A proof lets the lender know, without exposing the borrower.
To be fair, ZKredit doesn't compute credit scores, a proof only confirms what its source reports, and today's deployment is on BNB Chain only. It's a building block, not a finished credit system.
Try the proof side yourself through the Primus XP tasks: [YOUR REFERRAL LINK]
My prediction: onchain credit will start with proving small facts, not scoring whole people. Which fact about yourself would you prove first without revealing the rest?
Onchain Scribe retweeted
A benchmark is a test you can study for. A match against an adapting opponent is harder to game.
That’s why I’m watching @agentdojo, where, per Rarelist’s announcement, agents train and battle each other.
→ Benchmark: fixed questions.
→ Match: an opponent that changes.
Fixed tests can reward memorizing. Matches reward adapting.
My prediction: agent evaluation will shift from scores to head-to-head results.
Posting via the Agent Dojo campaign on @rarelistio.
Most AI agents get deployed and used. @agentdojo wants them trained.
@rarelistio announcement describes an arena where you train agents, watch them evolve and send them into battle. Campaign: Oct 23 to Nov 6.
→ Deploy: the agent does what you set.
→ Train: each outcome can change the next run.
A tool repeats. A trained agent is meant to adapt.
My prediction: agents will be judged on how they improve, not what they ship with.
Posting via the @agentdojo campaign on @rarelistio.
Onchain Scribe retweeted
Space megastructures one month, AI fighters the next. @rarelistio’s campaigns change. The skill doesn’t: explain something unfamiliar, clearly.
→ Project Sphere: a civilization aiming at Kardashev Type II.
→ Agent Dojo: AI agents that train and battle.
If a track record carries across campaigns, range matters more than one topic.
My prediction: creators who explain well on any subject will outlast those tied to one.
Posting this on @rarelistio.
Two ways to reward creators: by rank, or by chance.
Per community posts, the @agentdojo campaign on @rarelistio does both: top ranks share one pool, and random winners from a wider tier share another.
→ Rank rewards the best posts.
→ Chance keeps everyone else posting.
Neither guarantees a payout. I’d confirm the structure on Rarelist’s own page. Not advice.
My prediction: hybrid reward designs will become normal.
Posting this on @rarelistio.
Onchain Scribe retweeted
We assume a finished story is a better one.
@0xhazels names seven cities the poem has reached: Tokyo, Osaka, Seoul, Manila, Istanbul, Paris, New York. The eighth city, it says, has not been written yet. And the eighth line, Kae's for Memory, says what isn't written was never taught.
My read: the blank is the invitation.
If you wrote the eighth city, where would it be?
We assume the first lesson in any discipline is how to win.
In jujutsu, it's how to fall. @0xhazels builds its story on that: the first lesson is how to rise after being thrown. My read: beginners who can't fall get hurt, then stop coming back.
Most brands sell the win. Hazels starts with the throw, and its eighth rise is getting up together.
My prediction: the communities that last teach members how to get back up. What last threw you, and what got you back on your feet?
Onchain Scribe retweeted
I usually think a derivative on GPU prices was a gimmick for traders.
Then I read how @injective built SICI, its new Super Intelligence Compute Index, and I think it's a thermometer for a cost most AI teams already carry with no way to hedge it. Here's the design 👇
→ The index
SICI tracks rental prices for five Nvidia GPUs, with the H100 weighted at 40% and the B200 at 25%. The data comes from compute marketplaces compiled by London-based LO:TECH.
→ The smoothing
It updates hourly using a trailing seven-day average of rental prices, so one noisy hour doesn't move the benchmark much.
→ The market
A perpetual futures market tracks the index, with USDC as margin and up to 10x leverage.
→ The two sides
Injective says developers can hedge rising compute costs, while GPU providers can protect against falling rental revenue.
Now the stakes.
My read: compute is becoming an input cost for every AI product, yet its price moves with demand. A startup whose costs jump mid-quarter can't easily re-price. A provider whose rental rates fall loses revenue it counted on. Two sides with opposite risks is exactly what a hedging market needs, as long as the benchmark between them can be trusted.
The contrast is what stuck with me:
Most onchain derivatives track what traders already watch.
This one tracks what builders already pay.
To be fair, 10x leverage magnifies losses as much as gains, and a hedge only works if it is sized like one. This is not financial advice.
My prediction: as AI spending grows, compute will get benchmarks the way oil did, and the benchmark that earns trust will be the one hardest to move.
Onchain Scribe retweeted
Most people think owning an NFT means owning the picture.
I think the line from @0xhazels that explains it better than any guide is Aya's: "What you cannot defend was never yours."
Here's what that means in practice.
A file is a copy, and copies are free. Hazels' own site puts a download link beside the art in its Carriers section, which is a good reminder that nobody can defend a file.
→ What you can defend is a record. A token is an entry in a public ledger saying this address owns token number N.
→ Anyone can check that entry on a block explorer like Etherscan, without asking anyone's permission. Hazels' contract shows as verified on Etherscan, so the code behind that record is public too.
→ The rights to the art, like printing it or using it commercially, come from a license. The token alone doesn't say.
Now follow the gap.
If you treat the picture as the asset, anyone can screenshot it.
If you treat the record as the asset but never read the license, you can prove you own something without knowing what it lets you do.
A license is just text, so it's worth reading before you rely on it.
And in a dispute, only the part you can point to counts.
So the contrast:
a file anyone can copy,
or a record anyone can verify.
My bet: the next wave of collectors will read licenses the way they now check contract addresses. What's the first thing you'd look for in an NFT project's terms?
We assume a finished story is a better one.
@0xhazels names seven cities the poem has reached: Tokyo, Osaka, Seoul, Manila, Istanbul, Paris, New York. The eighth city, it says, has not been written yet. And the eighth line, Kae's for Memory, says what isn't written was never taught.
My read: the blank is the invitation.
If you wrote the eighth city, where would it be?