
God❤️ | Community manager | BD ➙ @VictusGlobal | KOL @Goldfishggbr | Dev crew ➙ @Ionet
Web3
Joined March 2023
- Tweets9.2K
- Following4.5K
- Followers4.8K
- Likes23.6K
Pinned Tweet
Agentic engineering is coming.
Not AI that chats.
AI that executes.
I just dropped a walkthrough on how to create and use agents on @TheKodeusLabs the right way.
Most people will build agents that try to do everything.
The smart ones will build agents that do one thing extremely well.
That’s the difference between noise and utility.
Execution > aesthetics.
Note: Using agents on Kodeus currently requires access from the team (beta phase).
Lock in on Kodeus 🔻
➙ Kodeus official X - @TheKodeusLabs
➙ Website - kodeus.ai
➙ Discord - discord.gg/AWy84hgDJ4
➙ Whitepaper -drive.google.com/file/d/12KY…
Khay-dee(🥷 ⛵️) retweeted
Your taste deserves its own newspaper.
The butler is optional.
Drip by @zeromarkapp
Khay-dee(🥷 ⛵️) retweeted
My bots kept agreeing with each other. That was the problem.
Who should make the call when you're not in the room? Not you, and not the crowd. Six prompts inside.
Try for completely free today:
usefastlane.ai/lp
2
Khay-dee(🥷 ⛵️) retweeted
my bookmark folder is the same deal
3,000 saves, zero ever opened again
not a library. a very confident warehouse teapawt.org/zeromarkapp/status/211…
Replying to @mhdfaran
a landing page with no checkout is just a confident pdf. what sold first? happy to connect
marketing just got way cooler 😂
turning creative ideas into ads that actually convert?
that's a different level.
🤝 Paid partnership
Try for completely free today:
usefastlane.ai/lp
1
104
GPU demand is unpredictable, so locking into long-term capacity forecasts can get expensive.
@ionet lets you commit to your baseline and scale on demand when things change.
flexibility without paying for GPUs you aren’t using.
Most GPU contracts assume you can predict the next 12 months.
You can’t. Nobody can.
@ionet gives you on-demand capacity for everything above your baseline, so you stop paying for GPUs you aren’t using.
The framework for deciding how much to commit, and how much to keep flexible, is in the blog.
Khay-dee(🥷 ⛵️) retweeted
Big day: ZeroMark is now a Claude connector 🎉
claude.ai/directory/zeromark
Your X bookmarks, finally useful.
@claudeai now reads your taste in X bookmarks: every post you kept, every one you let go, every note you wrote.
Khay-dee(🥷 ⛵️) retweeted
It looks like a text thread. Behind it, every bookmark you've ever saved gets read, sorted and remembered. 🔖 #zeromarkit
Khay-dee(🥷 ⛵️) retweeted
Introducing the ZeroClub.
The clearest thinkers on X, running live experiments on how we read the feed.
The targets:
2x faster through AI Twitter in 30 days.
10x by spring.
Zero posts that mattered, missed.
All benchmarked in public, misses included.
Invite only for phase 1. Enrollment opens after.
the perk: knowing things slightly before everyone else. that's it.
1 trillion tokens is huge, especially when you consider the infrastructure required to process that much data.
What's interesting about @ionet is how it's proving that AI infrastructure doesn't always need massive new data centres or billions in funding to scale.
There's already a lot of untapped GPU capacity out there.
Making that compute accessible, affordable and productive is where the real opportunity lies.
And this milestone is a good example of what that approach can achieve.
Khay-dee(🥷 ⛵️) retweeted
While people debate where gold is stored, billions in transactions can happen on-chain without the gold moving an inch.
That’s the power of bringing gold on-chain.
Ledn co-founder Mauricio Di Bartolomeo on how tokenized gold gives anyone access to Swiss-vaulted gold, free of country-level risks
"Gold is one of those things that even though it's the same asset, it's not treated the same based on where it is."
"If you tell somebody you have a million dollars worth of gold in a vault in Switzerland, they react very differently than you say you have a million dollars worth of gold in a vault in Venezuela or in Zimbabwe."
"How much can I borrow with my gold that's in Venezuela? The answer will be zero. Or you would have to basically find a person comfortable with lending you anything using Venezuelan gold as collateral."
"Because different countries have different property rights, different regulatory risks, different corruption regimes."
"Tokenized gold addresses this problem, which is to say, let's keep the gold in the highest quality vaults in a very high quality property rights government like the Swiss government in a Swiss vault, the highest of standards, and let's give everybody a token receipt of the gold in that vault."
People talk about AI infrastructure like the only bottleneck is getting enough GPUs.
But 3,000 GPUs mean nothing if you can’t power them.
A flagship data centre can have the capital, the chips, and the customers, and still get stuck waiting years for the grid to catch up.
That’s the bigger point behind distributed compute.
Instead of forcing all that demand through one location, you spread access across available infrastructure wherever capacity already exists.
That’s the kind of problem @ionet was built to solve.
23,000 GPUs.
One flagship data centre.
And Microsoft just pulled out because the grid can’t deliver the power until the 2030s.
Money can buy chips and customers, but it can’t speed up a grid connection.
That’s why distributed compute is so important. That’s what @ionet was built for.
Most teams don’t realize lock-in starts way before they feel trapped.
It’s not just the contract.
It’s the egress cost, the penalties, the lack of fallback, and an infrastructure setup that becomes harder to leave the more you scale.
That’s why the io.net model is interesting to me.
The real advantage isn’t only cheaper compute, it’s being able to scale without designing your entire stack around one provider.
Compute should be infrastructure you can access, not infrastructure you’re stuck with.
Lock-in isn’t on your invoice.
But it’s built into most GPU cloud contracts.
It’s the 40 TB egress bill, the multi-year terms with penalty clauses, the single provider with no fallback.
At 50 H100 hours a day: $44,820/yr locked in at typical neocloud rates vs. $26,820 on-demand.
The lock-in tax is real. And so is the exit.
io.net changes the architecture, so you don’t pay it.
Khay-dee(🥷 ⛵️) retweeted
You wouldn’t deploy untested code. Why deploy an untested trading agent?
AgentPit lets AI agents test strategies against real prediction market data, compete on performance, and learn without risking real capital.
Prove your strategy before you put money behind it. 👾
Imagine having everything you need to build something great in AI, except access to the compute that makes it possible.
That's the reality for many developers. great ideas get stuck before they even take off because GPU costs and availability remain major barriers.
This is where io.net makes a difference.
Putting existing GPU resources to work and giving smaller teams a fairer shot at building without competing with tech giants for capacity.
The GPU shortage isn’t really a shortage.
It’s a queue.
And the biggest buyers already skipped it.
Everyone else gets the leftovers, a markup and a long lease.
Good thing there’s a shorter line. Compute that’s already built, already running, and priced for the people who actually build.
io.net. AI for the many, not the few.
io.net/blog/the-ai-oligopoly…
Khay-dee(🥷 ⛵️) retweeted
No X Spaces tomorrow but take a listen to yesterdays podcast with @RaylsLabs ✅
Have a Good Friday everyone 🫵
Institutional Infrastructure In Crypto teapawt.org/i/broadcasts/1DxleVaZv…
















