Pinned Tweet
🧵1/8 How I went from $1,000 → $100,000, then back down to $0 and now eventually to 7 figures 👑
Started as a college student with $1,000, I had no experience, just willingness to take risks
I had no idea crypto would completely change my life
Privacy just got easier
@Nymrasi app is LIVE!
You can send cash out & get paid privately
It's now live on base, robinhood, ethereum, and bnb
You can also scan any EVM or Solana wallet to see what your activity reveals
Want to try it? Link is below 👇
The Nymra App is live.
Go private, send, cash out and get paid from your own private wallet. Only you see what it holds.
- Live on Robinhood Chain, Base, Ethereum and BNB Chain
- Privacy Scan for any EVM or Solana wallet
nymra.si/app
🚨 JUST IN: Reddit $RDDT crashes nearly 30% in 24 hours!
Google AI search is eating into Reddit referral traffic and fears of slowing user growth plus weaker Q4 guidance have sent investors running
Why multi-angle captures matter
One camera angle cannot always show the full shape of a place
Different views can reveal details that a single recording might miss
@vangrid_io capture approach helps turn footage of real environments into 3D models
Capturing a space carefully from multiple angles can help provide a more complete view of the surroundings
Good data starts with good capture habits
🤝 Paid partnership
Anyone can earn through @vangrid_io bounties
There are plenty of bounties available, from showing manholes and roads to capturing useful real-world data.
Simply complete bounties and get rewarded for your contributions.
So far, over $669K has been settled in rewards.
Your surroundings could be worth more than you think.
🤝 Paid partnership
😂 Binance founder Changpeng Zhao (CZ) has threatened to sell his $SPCX shares to cover his losses
This comes after Elon Musk repeatedly trolled CZ
Replying to @elonmusk @HuaweiMate6
Elon, have a lot of respect for you, but bad time to troll. Please stop.
Our SPCX shares (from our $500m investment in your Twitter takeover) are worth more than a billion.
Would appreciate if you show a little support. Thanks!
I think physical AI needs better coverage
A robot working inside a building faces different challenges from a robot moving outdoors
It needs to understand
- Obstacles
- Spaces
- Routes
- And the layout around it
@vangrid_io aims to expand real-world data coverage by collecting captures from different locations through smartphones
More varied environments mean more opportunities to build a useful picture of the physical world
🤝 Paid partnership
Have you guys wonder why bounty-based data collection is interesting
Imagine a company needs data from a particular hallway or warehouse
Instead of relying only on a large mapping team it could create a specific capture request
@vangrid_io bounty system provides a way to connect those requests with people who can collect the footage
It is an interesting approach to making real-world data collection more flexible
🤝 Paid partnership
🚨 X links have an interesting new detail that people should look at
When you copy a post link, you might notice something like this at the end
?a=D4O-SC9JXzDbxBdjrKk&s=20
?a=
s=20
What exatly mean of these?
How you can earn through @vangrid_io
So vangrid has a bounty system where people can capture requested locations and submit footage for review
When a submission is accepted the associated reward can be paid in USDC according to the bounty terms
The reward depends on the task and whether your capture is selected
So always check the requirements before participating
🤝 Paid partnership
It's actually very insane
Google Gemini 4 Carbon is beating current top models, including Claude Opus 5.5 and Gemini Argon 4
It's a top AI model in cybersecurity
You wont believe how good these 2 things are: zkFHE and zkTLS by @primus_labs
Let me explain them in the simplest way possible
=> zkTLS
helps you prove that information from a website is real without exposing all your private data
Imagine proving your Binance trading results without showing everyone your entire account that's the kind of thing zkTLS can help make possible
=> zkFHE
is about doing calculations on encrypted data without needing to reveal the original information think of it like solving a maths problem while keeping the numbers hidden
So what is Primus working toward with these technologies?
Making online data more verifiable while protecting privacy opening the door to applications in DeFi trading verification, AI, and other Web3 use cases
=> Why is this good?
Users can prove more while exposing less and developers can build applications that don't have to rely blindly on screenshots or trust-based claims
Privacy and verifiable data are going to matter a lot as Web3 and AI grow
🤝 Paid partnership
Have you guys wondered what makes real-world data valuable?
@vangrid_io focuses on capturing real places and turning that footage into data that can support Physical AI
Different environments bring different challenges, from narrow indoor spaces to busy outdoor areas
The goal is to give intelligent machines a better understanding of the world they may one day work in
🤝 Paid partnership
I know everyone loves vibe coding these days but the real pain is still there
You sit and watch the screen while the agent keeps asking for permission one by one
Then CI fails and you have to jump back in again
@SuperPlanehq just launched 1.0
It is an open source software factory for engineering teams that handles this coordination for you
It scans the
-> backlog
-> picks the right tickets
-> runs the agent
-> works through the CI failures
-> and only brings the result to you when it is ready for review
That is the part most of us still do by hand
🤝 Paid partnership
Last month, 52% of the pull requests our team merged were automated through SuperPlane.
Today, we’re launching @superplanehq 1.0: an open source software factory for engineering teams.
We built it because using a coding agent can feel like becoming its assistant.
Starting sessions. Checking its work. Chasing failing CI.
SuperPlane takes on that coordination:
- Scans your backlog and proactively picks up tickets suited to automation.
- Runs a coding agent and verifies the result.
Works through CI failures and review feedback.
- Your team reviews the evidence and decides what ships.
The setup, instructions, and standards are shared across the team. When someone improves the process, everyone benefits.
Choose your agents and models. Run it on your own infrastructure.
We spent more than a decade building CI/CD at Semaphore. We’ve seen how much teams can accomplish when routine work runs reliably in the background.
Now we’re bringing that approach to coding agents.
Here’s how it works:
RIPPLE JUST GOT FULL FEDERAL APPROVAL FOR ITS TRUST BANK
What does this actually mean?
Think of it as another layer of government oversight for Ripple's crypto business
-> It can manage reserves backing RLUSD
-> It can provide digital asset custody for businesses
-> It adds federal supervision alongside existing New York state oversight
Important: This doesn't mean ripple:native will automatically rise
But it could help Ripple build more trust with institutions
Crypto and traditional finance are getting closer



















