Web3 & Blockchain Ecosystem Builder | Community Growth Lead @BlockSurgeHQ | Simplify DeFi l Fogo Believer |
UAE
Joined May 2025
- Tweets43.1K
- Following2.1K
- Followers3.1K
- Likes80.6K
Pinned Tweet
GM CT Fam
Launching a 60,000 $FOGO community challenge around the @BaranosAI Educational Quest. 🔥
The 10 best content creators will receive 6,000 $FOGO each, with six days to enter.
I have built the Quest to make the Baranos whitepaper easier to understand through a game. You work through the cases, develop your own use case and earn a certificate along the way.
I’ve always believed onchain AI has a big future. Baranos on-chain verification of AI is really quite interesting and verify the computation behind an AI decision. Once smart contracts act on those results, that ability starts to matter a lot.
Now I’d love to see how you explain it.
To enter:
1. Complete the Quest and submit your use case to unlock your certificate.
2. Read the whitepaper and create original content about what you learned.
3. Quote-post this announcement with your content and certificate, tagging @BaranosAI and @fogo.
A post, thread, graphic or short video all count.
Explain why Baranos’s approach matters, or explore something useful you’d build with it on Fogo’s 40ms block infrastructure. Tell us who it helps, why verification matters and what still needs work. Be clear about what exists today and what you’re proposing.
I’ll look for originality, clear explanations, technical accuracy and practical ideas when selecting the winners.
Entries will close on Saturday 5th October 2026.
"Play the Quest" baranosaieducationalquest.ne…
"Read the whitepaper" baranos.ai/assets/baranos-wh…
P.S. This is an independent, community-led event, not organised by either the Fogo or BaranosAI team.
Dear Ritualists! 💚
We all know @ritualnet Mainnet is getting closer and we can already feel the community energy picking up
50M $RITUAL is officially allocated to the community airdrop at launch
I built Ritual Airdrop Playground for the community to play with some mental maths *pure-preMainnet fire vibes*😂
Curious what your Ritual roles could translate into?
Pick your roles, tweak the assumptions, move the price around and see how different scenarios play out
No wallet connection or claiming, also its not the official allocation
just intend to create community curiosity, a bit of maths and plenty of Ritual hopium while we wait for the real thing
I’ve spent a lot of time learning, creating and building around Ritual, so making something the community can actually interact with felt much more fun
Ritual season is starting to feel very real now
Check it out and see what your expected airdrop could look like:
whimsical-marigold-50a949.ne…
Curious to know what does your hypothetical Ritual Airdrop looks like? 👀✌️
Good Luck, Ritual Fam ❖
Share your Ritual Hypothetical Airdrop Scenario Card for fun.
🤖 Made with AI
We spend so much time in crypto protecting seed phrases, checking approvals and avoiding wallet drains.
But at some point, the bigger security question may sit much deeper than that.
It may be the cryptography securing the wallet itself.
As quantum computing keeps progressing, crypto will eventually have to move away from some of the signature systems we use today and toward post-quantum cryptography, or PQC.
Then @toly raised a point that really triggered me and I thought about the next step too.
What happens if the PQC scheme we move to later gets broken?
Maybe quantum forces us to replace today’s cryptography, but then AI becomes good enough to find weaknesses in the new one much faster than we expect.
That is why I like Toly’s idea of redundancy.
Instead of trusting one new “quantum-safe” signature, imagine a wallet using 3 different signature schemes and requiring any 2 of them to move funds.
If one gets compromised, you still have another layer protecting the wallet.
For me, that feels much safer than replacing one lock with another lock and assuming the problem is solved.
PQC itself should not become the next single point of failure.
Quantum may change the locks.
AI may become very good at testing those locks.
So the smarter design could be making sure one broken lock is never enough to take the wallet down.
I think we are going to hear a lot more about this side of wallet security.
woman defi retweeted
This is what we were waiting for, exciting times ahead 😄
Update on @BaranosAI ?
GPU’s are hard to come by these days – and having to ship and clear customs in Japan takes time.
Good news is that’s now done, and it’s racked and stacked in @TheHUB
Stay tuned for initial use-case.
Security first.
This is what we were waiting for, exciting times ahead 😄
Update on @BaranosAI ?
GPU’s are hard to come by these days – and having to ship and clear customs in Japan takes time.
Good news is that’s now done, and it’s racked and stacked in @TheHUB
Stay tuned for initial use-case.
Security first.
woman defi retweeted
GM CT Fam
Happy Weekend
Lets understand this, we allow AI to take a few hours of paperwork off our hands regularly. Giving it permission to pay someone is where we start asking more questions.
Say you run a small business and your AI checks invoices against signed contracts before approving payments. That could save you hours every week.
You’ve chosen the model, the documents it should use and the instructions it should follow. Then the service quietly switches to a cheaper model or changes those instructions.
You still get a payment confirmation. From your side, everything looks normal. Yet the process you agreed to has changed.
Its the narrative @BaranosAI brings forward.
Verifiability!
Its design commits each job to a specific model, inputs and execution rules. Independent participants can reproduce the computation and challenge incorrect execution, with disputed steps replayed and adjudicated on @fogo
In that business scenario, an app could wait for the settled result before releasing funds. The owner wouldn’t need to inspect every calculation themselves.
There would still be things to check.
The model could misunderstand a contract, or someone could submit a forged delivery record. Verifying the computation doesn’t make the evidence truthful or the decision sensible.
But it gives the application a way to check whether the agreed computation was followed.
I can see AI giving small teams time back and helping them take on work they couldn’t manage before. So we are ready to give those systems more responsibility, I’d want verification built into the apps we use, especially the ones allowed to move our money.
Keep an eye on @BaranosAI and see the AI revolution being solidified.
GM CT Fam
Happy Weekend
Lets understand this, we allow AI to take a few hours of paperwork off our hands regularly. Giving it permission to pay someone is where we start asking more questions.
Say you run a small business and your AI checks invoices against signed contracts before approving payments. That could save you hours every week.
You’ve chosen the model, the documents it should use and the instructions it should follow. Then the service quietly switches to a cheaper model or changes those instructions.
You still get a payment confirmation. From your side, everything looks normal. Yet the process you agreed to has changed.
Its the narrative @BaranosAI brings forward.
Verifiability!
Its design commits each job to a specific model, inputs and execution rules. Independent participants can reproduce the computation and challenge incorrect execution, with disputed steps replayed and adjudicated on @fogo
In that business scenario, an app could wait for the settled result before releasing funds. The owner wouldn’t need to inspect every calculation themselves.
There would still be things to check.
The model could misunderstand a contract, or someone could submit a forged delivery record. Verifying the computation doesn’t make the evidence truthful or the decision sensible.
But it gives the application a way to check whether the agreed computation was followed.
I can see AI giving small teams time back and helping them take on work they couldn’t manage before. So we are ready to give those systems more responsibility, I’d want verification built into the apps we use, especially the ones allowed to move our money.
Keep an eye on @BaranosAI and see the AI revolution being solidified.
woman defi retweeted
Fogo in Nearly a Year
From a Fast Blockchain to New Infrastructure for Onchain Finance.
If we look beyond first layer DeFi, the most interesting thing about Fogo right now isn’t another protocol. It’s the shift in what this network could offer the market.
I see three directions worth watching.
@TheHUB the Physical Layer.
Colocation infrastructure in Equinix data centers in Tokyo, New York, and London. The idea is to reduce network latency and narrow the advantage of participants who simply have a shorter path to trading infrastructure.
TheHUB isn’t an exchange it’s a physical environment designed to make trade execution more predictable and this is where @fogo moves beyond the usual conversation about block speed.
@slx_fi new Approach to Trade Execution.
Prism and Dual Flow Batch Auctions aim to change a system where the advantage goes to whoever gets their order in first. Instead, market makers compete to offer traders the best price.
Your price or better. Every trade.
This is where Fogo’s technology could find a practical use case however with Superluminal still in private testing, the real proof lies ahead: live trading, liquidity, and execution quality at scale.
@BaranosAI verifiable Intelligence.
I find this direction interesting for reasons that go beyond the word AI.
Baranos aims to make LLM outputs verifiable by recording the model, input data, evidence, and execution rules, while allowing disputed computations to be reproduced onchain.
This opens up interesting possibilities for applications where AI outputs influence the movement of funds or determine the outcome of an event. Superluminal has already publicly explored prediction markets and verifiable AI-based settlement as potential directions for development.
If these components can be brought together, we could get something more interesting than just another L1: infrastructure for trading, an order execution mechanism, and a way to verify the outputs that financial decisions depend on.
But there’s another side to this.
The market isn’t forgiving of weak liquidity right now. Capital is much more willing to concentrate in the largest assets, while altcoins remain in a more difficult position. FOGO is approximately 94% below its ATH, and according to DeFiLlama data as of October 23, Fogo’s spot DEX volume was only around $1.9K over 24 hours.
This isn’t just a consequence of the market cycle. It’s also a signal that technological ambitions haven’t yet translated into sufficient organic demand.
That’s why I’m curious to see what comes next.
Can Superluminal prove its model in a live market?
Will TheHUB become a tangible competitive advantage?
Will Baranos AI evolve from a technical demonstration into a tool that other applications actually use?
For me Fogo’s next phase isn’t about proving how fast a blockchain can be. It’s about whether these three directions can work together and generate the demand the ecosystem still lacks.
woman defi retweeted
Happy Friday y'all 💚
The dominating AI compute is not one-size-fits-all
Extremely important notion that a small inference, a huge reasoning model and a private verified job should not all be sent to the same machine under the same rules
so @ritualnet Resonance work makes so much sense
its that simple idea of matching the job with the right compute
Hardware, models, latency, prices and guarantees may be different
MLCommons benchmarks already show how dramatically AI workloads differ in compute and performance requirements
Ritual takes that reality and solves question like
what if compute itself could become an open market?
giving a much closer feel to how AI infrastructure actually needs to work
Volkswagen passed emissions tests while its cars were polluting the environment
The software recognised when it was being tested and changed its behaviour. In 2015, regulators uncovered the deception. Affected cars could emit nitrogen oxides at up to 40 times the permission.
I think about that when we talk about trusting AI with money.
Volkswagen wasn’t an AI incident. But it showed how much can go wrong when people cannot inspect the software behaviour behind a reassuring result.
That’s where @BaranosAI has an edge.
Its design commits to those details before the computation runs. Independent participants can reproduce the job, challenge an incorrect execution and have the disputed computation checked onchain through @fogo
Suppose a prediction market utilizes AI to assess evidence. If the operator quietly substitutes another model or changes the inputs, participants need a way to challenge that before accepting the settlement.
Of course, the agreed model can still be wrong. Verification won’t fix poor judgment. It checks that the agreed computation was followed.
Learning more n more about BaranosAI made this much more tangible for. The more responsibility we give AI, the more valuable that ability to check and challenge becomes.
Happy Friday y'all 💚
The dominating AI compute is not one-size-fits-all
Extremely important notion that a small inference, a huge reasoning model and a private verified job should not all be sent to the same machine under the same rules
so @ritualnet Resonance work makes so much sense
its that simple idea of matching the job with the right compute
Hardware, models, latency, prices and guarantees may be different
MLCommons benchmarks already show how dramatically AI workloads differ in compute and performance requirements
Ritual takes that reality and solves question like
what if compute itself could become an open market?
giving a much closer feel to how AI infrastructure actually needs to work
Happy Friday y'all 💚
The dominating AI compute is not one-size-fits-all
Extremely important notion that a small inference, a huge reasoning model and a private verified job should not all be sent to the same machine under the same rules
so @ritualnet Resonance work makes so much sense
its that simple idea of matching the job with the right compute
Hardware, models, latency, prices and guarantees may be different
MLCommons benchmarks already show how dramatically AI workloads differ in compute and performance requirements
Ritual takes that reality and solves question like
what if compute itself could become an open market?
giving a much closer feel to how AI infrastructure actually needs to work
woman defi retweeted
Volkswagen passed emissions tests while its cars were polluting the environment
The software recognised when it was being tested and changed its behaviour. In 2015, regulators uncovered the deception. Affected cars could emit nitrogen oxides at up to 40 times the permission.
I think about that when we talk about trusting AI with money.
Volkswagen wasn’t an AI incident. But it showed how much can go wrong when people cannot inspect the software behaviour behind a reassuring result.
That’s where @BaranosAI has an edge.
Its design commits to those details before the computation runs. Independent participants can reproduce the job, challenge an incorrect execution and have the disputed computation checked onchain through @fogo
Suppose a prediction market utilizes AI to assess evidence. If the operator quietly substitutes another model or changes the inputs, participants need a way to challenge that before accepting the settlement.
Of course, the agreed model can still be wrong. Verification won’t fix poor judgment. It checks that the agreed computation was followed.
Learning more n more about BaranosAI made this much more tangible for. The more responsibility we give AI, the more valuable that ability to check and challenge becomes.
GM frens☀️
Happy Thursday
pretending AI behaves like deterministic EVM code is not something to be bullish on
AI is asynchronous, probabilistic and occasionally messy and so features like callbacks, TTLs, explicit failure states, sender verification and output parsing are some treasure features
and @ritualnet, doing dev plumbing secures AI failures onchain
p.s. loved building on ritual onchain as well as offchain
woman defi retweeted
GM CT 💚
If an AI agent disappears the moment one server goes down, I’d struggle to call that fully autonomous
The @ritualnet persistent-agent design feels important here
Agents getting better at reasoning and taking actions, but long-running reliability is still a problem
Research on agentic systems keeps showing the same thing like being capable is not the same as being dependable over time
Ritual is dope tackling the infrastructure side
Agents can save its state, prove it is still alive, and recover from its last checkpoint just so unlike dying with the machine that hosted it
For something managing capital, running a strategy, or coordinating with other agents, that matters a lot, what do you say?
GM CT 💚
If an AI agent disappears the moment one server goes down, I’d struggle to call that fully autonomous
The @ritualnet persistent-agent design feels important here
Agents getting better at reasoning and taking actions, but long-running reliability is still a problem
Research on agentic systems keeps showing the same thing like being capable is not the same as being dependable over time
Ritual is dope tackling the infrastructure side
Agents can save its state, prove it is still alive, and recover from its last checkpoint just so unlike dying with the machine that hosted it
For something managing capital, running a strategy, or coordinating with other agents, that matters a lot, what do you say?
woman defi retweeted
Building the BaranosAI Educational Quest left me with a question I keep coming back to:
Can an ordinary participant actually afford to hold an AI executor accountable?
One detail in @BaranosAI design deserves more attention. Its Durable Compute Graph breaks computation into small, deterministic operations called kernels. A dispute progressively narrows the disagreement until the chain can adjudicate one kernel, which fits inside a single SVM transaction. The full challenge still involves multiple rounds.
That gives Baranos a promising way to limit the computation needed for the final onchain ruling.
But affordability needs the whole bill.
A watcher still recomputes the result offchain and must respond within deadlines. The developer docs list a 1,000,000-lamport challenger bond in the alpha tooling. That is configurable capital at risk, not the price of verification. They also estimate roughly an hour of executor work for a 10,000-token document, which shows why hardware and workload matter alongside gas.
I haven’t found a reproducible benchmark that puts the complete challenger cost into dollars. The whitepaper itself calls for total chain, execution and verification costs, with hardware, storage and network conditions disclosed.
That’s the next demonstration I’d love to see: an independent participant challenging a deliberately incorrect execution, with the compute bill, transaction costs, bond and time to settlement all published.
If those economics work for smaller participants, Baranos could make accountability much more accessible. That would give builders on @fogo a strong reason to use it.
After helping people learn how verification works, I want to see how affordable it is to exercise that right.
🤝 Paid partnership
Building the BaranosAI Educational Quest left me with a question I keep coming back to:
Can an ordinary participant actually afford to hold an AI executor accountable?
One detail in @BaranosAI design deserves more attention. Its Durable Compute Graph breaks computation into small, deterministic operations called kernels. A dispute progressively narrows the disagreement until the chain can adjudicate one kernel, which fits inside a single SVM transaction. The full challenge still involves multiple rounds.
That gives Baranos a promising way to limit the computation needed for the final onchain ruling.
But affordability needs the whole bill.
A watcher still recomputes the result offchain and must respond within deadlines. The developer docs list a 1,000,000-lamport challenger bond in the alpha tooling. That is configurable capital at risk, not the price of verification. They also estimate roughly an hour of executor work for a 10,000-token document, which shows why hardware and workload matter alongside gas.
I haven’t found a reproducible benchmark that puts the complete challenger cost into dollars. The whitepaper itself calls for total chain, execution and verification costs, with hardware, storage and network conditions disclosed.
That’s the next demonstration I’d love to see: an independent participant challenging a deliberately incorrect execution, with the compute bill, transaction costs, bond and time to settlement all published.
If those economics work for smaller participants, Baranos could make accountability much more accessible. That would give builders on @fogo a strong reason to use it.
After helping people learn how verification works, I want to see how affordable it is to exercise that right.
🤝 Paid partnership
woman defi retweeted
GM CT 🌻
I absolutely loved reading this...
dealing with the part most AI x crypto narratives skip; failure handling
In production, AI is not a clean synchronous function call means responses can be late, malformed, fail mid-job, or come back through the wrong path
The model @ritualnet's callback forces the contract to account for that explicitly
in terms of timing, sender verification, execution success or output parsing
much more mature design philosophy I'd say
The real infrastructure edge is surely making AI callable and making AI failure legible, bounded & safe enough for contracts to rely on 💚
🤝 Paid partnership
GM CT 🌻
I absolutely loved reading this...
dealing with the part most AI x crypto narratives skip; failure handling
In production, AI is not a clean synchronous function call means responses can be late, malformed, fail mid-job, or come back through the wrong path
The model @ritualnet's callback forces the contract to account for that explicitly
in terms of timing, sender verification, execution success or output parsing
much more mature design philosophy I'd say
The real infrastructure edge is surely making AI callable and making AI failure legible, bounded & safe enough for contracts to rely on 💚
🤝 Paid partnership