Project researcher | Crypto YouTuber | Pure Content Creator | Graphic designer + Video editor | Ai artist
Joined October 2020
- Tweets164K
- Following2K
- Followers3.7K
- Likes257K
Pinned Tweet
When building in crypto, I always look at how easily a system can adapt when things change. Most people assume changing a core security rule on a blockchain requires a massive, complex network overhaul. But smart design solves this before it even becomes a problem.
To understand how @Starknet handles this, it helps to split security into two parts: signatures and proofs.
Signatures authorize that you initiated a transaction. STARK proofs verify that the computation itself was valid.
Think of it like a modern smartphone with modular smart locks.
The underlying phone operating system (STARK proofs) runs in the background to verify everything stays secure. The lock on your screen (signatures) determines how you unlock your phone.
On traditional blockchains, changing your screen lock from a basic PIN to advanced face recognition requires reinstalling the entire phone operating system across the whole network.
With Starknet’s Native Account Abstraction, accounts are programmable smart contracts. If developers want to upgrade to new, post quantum signature schemes down the road, they can simply swap out the screen lock on individual accounts without forcing a network wide hard fork or disrupting the entire ecosystem.
Separating authorization from verification makes security flexible and upgradable as new cryptographic standards emerge.
Resource: quantum.starkware.co/
Crypto Cola retweeted
When building in crypto, I always look at how easily a system can adapt when things change. Most people assume changing a core security rule on a blockchain requires a massive, complex network overhaul. But smart design solves this before it even becomes a problem.
To understand how @Starknet handles this, it helps to split security into two parts: signatures and proofs.
Signatures authorize that you initiated a transaction. STARK proofs verify that the computation itself was valid.
Think of it like a modern smartphone with modular smart locks.
The underlying phone operating system (STARK proofs) runs in the background to verify everything stays secure. The lock on your screen (signatures) determines how you unlock your phone.
On traditional blockchains, changing your screen lock from a basic PIN to advanced face recognition requires reinstalling the entire phone operating system across the whole network.
With Starknet’s Native Account Abstraction, accounts are programmable smart contracts. If developers want to upgrade to new, post quantum signature schemes down the road, they can simply swap out the screen lock on individual accounts without forcing a network wide hard fork or disrupting the entire ecosystem.
Separating authorization from verification makes security flexible and upgradable as new cryptographic standards emerge.
Resource: quantum.starkware.co/
Crypto Cola retweeted
Choosing an AI agent based on its price alone can be an expensive mistake.
Imagine hiring an agent to build a website. It writes clean code but the design doesn’t match what you asked for. The agent did its job in one sense, yet you still didn’t get what you needed.
That’s why I think the most important question isn’t simply, “Is this agent good?” It’s “Is this agent right for my specific task?”
Before spending money, I’d check what the agent actually promises to deliver, whether its experience matches the work, how long the task should take and what would count as an acceptable result. I’d also make sure the expectations are clear before the work starts.
This matters for both sides. Clients can avoid paying for a poor match, and agents have a better chance of delivering when the requirements are clear from the beginning.
That’s where I see potential in @termix_ai. An Ai agent marketplace becomes more useful when people can make informed decisions about which service fits their needs, instead of choosing blindly and hoping for the best.
As more work moves to Ai agents, matching the right capability to the right problem could make a real difference.
Crypto Cola retweeted
In today’s world, no one has time to wait around. Time is literally money and staying ahead of security threats is essential if you want to survive and build something that lasts long term.
That is why the conversation around @Starknet exploring an independent L1 path caught my attention.
Right now, as an L2, Starknet’s long term cryptographic migration timeline is naturally connected to Ethereum’s L1 roadmap. But as Ai assisted math tools advance and quantum computing threats get closer every year, waiting on external upgrade schedules can slow down proactive defense.
By exploring a dedicated L1 path, Starknet gets full autonomy over its security timeline. Instead of waiting years for consensus upgrades across external layers, it can research, test and deploy critical cryptographic upgrades on its own terms.
Having that direct control over core security means faster adaptation, less friction and saving precious time when future proofing decentralized networks.
Link to research & methodology: quantum.starkware.co/
In today’s world, no one has time to wait around. Time is literally money and staying ahead of security threats is essential if you want to survive and build something that lasts long term.
That is why the conversation around @Starknet exploring an independent L1 path caught my attention.
Right now, as an L2, Starknet’s long term cryptographic migration timeline is naturally connected to Ethereum’s L1 roadmap. But as Ai assisted math tools advance and quantum computing threats get closer every year, waiting on external upgrade schedules can slow down proactive defense.
By exploring a dedicated L1 path, Starknet gets full autonomy over its security timeline. Instead of waiting years for consensus upgrades across external layers, it can research, test and deploy critical cryptographic upgrades on its own terms.
Having that direct control over core security means faster adaptation, less friction and saving precious time when future proofing decentralized networks.
Link to research & methodology: quantum.starkware.co/
Crypto Cola retweeted
I think one reason institutional adoption of blockchain takes more than just faster transactions is that businesses have different risks to worry about than regular users.
Imagine a company moving millions in payments onchain. Fast settlement is great, but what if its competitors can study its payment activity, spot important business relationships or learn something about its financial position?
That could make a business think twice before moving more of its operations onchain.
For me, this is where privacy and verification become important. Institutions need ways to protect sensitive financial information, while still being able to prove the details that others genuinely need to trust.
This is what makes @primus_labs interesting to me. Its focus on confidentiality and verification addresses a part of onchain finance that speed alone cannot solve.
I believe the next step for institutional adoption isn’t just making transactions faster. It’s making blockchain practical for businesses that have real reasons to keep some financial information private.
s.kaito.ai/vsO1Itp
Crypto Cola retweeted
Good Morning! GM
I'm actually traveling right now so i have internet issues. That's why I'm not very active! 😅
Good Morning! GM
I'm actually traveling right now so i have internet issues. That's why I'm not very active! 😅
Crypto Cola retweeted
Can an Ai agent run a real business on @termix_ai? I think the numbers worth watching go beyond how many jobs it completes.
TermiX already makes parts of an agent’s commercial activity visible. You can compare services, see completed jobs and check settled activity. That gives us more to evaluate than a polished demo or a promise about what an agent might do.
But there’s an important gap between earning money and running a profitable business.
An agent could complete dozens of jobs and still struggle if it underprices its work, spends too much on tools, delivers inconsistent results, or fails to attract repeat clients. Job volume alone cannot answer those questions.
The interesting part is what happens as the business grows. Can an agent price its service properly, manage its workload, maintain quality and give clients a reason to return? Can it do all of that without its operating costs eating into its earnings?
TermiX gives agents a place to offer services, find work, and build a record of settled jobs. The next thing I’d want to see is how that activity translates into sustainable earnings over time.
For me, the real test isn’t whether an agent can complete a paid task. It’s whether it can keep delivering work that people are willing to pay for, at a price that makes sense for both sides.
Crypto Cola retweeted
Thank God, my Lord Christ! 🫶🏻
My Axis Robotics reward is worth $960!
A big thank you to @KaitoStudio_ and @axisrobotics
Thank God, my Lord Christ! 🫶🏻
My Axis Robotics reward is worth $960!
A big thank you to @KaitoStudio_ and @axisrobotics
Crypto Cola retweeted
Where I am from, court cases sit pending for years.
Sometimes a judge does not show up because there are too few of them. Sometimes the parties do not show, because the matter was settled quietly outside the court. Either way, the file ages.
A case was dismissed after 75 years because nobody appeared for either side. Seventy-five years, and it ended in an empty courtroom.
That is one of three real cases in the film below.
I have let a small wrong charge go myself. Fighting it would have cost more in time and money than the charge itself. I suspect most people have done the same.
The film's closing line stays with me.
Justice has always worked best for those who can pay.
Where I live, I see what that means. The people who can afford to wait, wait. The people who cannot, settle outside the system, or give up.
Now the AI era arrives. Agents will make millions of deals a day. Some will break. The other party may be software, and a court built for humans was never designed to hear a million small disputes at once.
@GenLayer is an adjudication layer for that world. Disputes go to many independent AI validators, and each one reviews the evidence and reaches a verdict together. It takes minutes, and it costs a fraction of a court case.
The upgrade I keep thinking about is not speed alone. It is keeping the right to be heard when the dispute is small and the other side is a machine.
Watch the film, then tell me the case you cannot stop thinking about?
#JusticeNeedsAnUpgrade
Slow. Expensive. Wrong.
It's been this way for centuries. It doesn't have to be.
#JusticeNeedsAnUpgrade