Aiden (@aidendeng_eth) header
Aiden (@aidendeng_eth) profile picture

Web3 Artist | Content/Contributor @liquidtrading

Joined August 2021
Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)Photo by Aiden (@aidendeng_eth)
The farm is taking shape. 🌾 7,777 Farmers. One Living World. This is how HypeFarmers works: gameplay, progression, trading, seasons, and Native HYPE rewards. Welcome to the farm.
584
560
582
912
55,112
Whoever guesses how many buttons are here gets a WL spot (I’m making the intern count all of them lol)
798
612
12
2,131
55,878
gLiquid I've been staring at the chart for too long... Feel like longing today #axel #liquidtrading
9
19
224
Next stop on the @liquidtrading world tour: Vietnam! 🇻🇳 Putting Axel in our traditional Ao Dai was a must, but I wanted to keep it modern and authentic. I made sure to include the details that really make it home for me the Non Quai Thao, the ceremonial offerings, and of course, the national flower. Hope you guys love this look as much as I do! #axel #liquidtrading
10
1
1
25
606
Perle Labs post-snapshot checklist. What’s dropping FIRST? 👀 Season 1 is officially locked Points secured on the leaderboard $PRL just hit the Coinbase roadmap And the team literally said “Details on contributor rewards and the next chapter for Perle are coming soon.” Here’s my predicted order in the next 2 weeks - Tokenomics reveal + $PRL utility breakdown - Rewards/claim portal launch - Points → $PRL conversion mechanics - Season 2 kickoff $17.5M funding, top-tier VCs, real product live… this could be the cleanest execution we’ve seen in AI x crypto in a while. $PRL @PerleLabs
2
1
1
10
301
$PRL is coming soon Asset added on Coinbase often takes two weeks to be listed. See you guys at TGE! #PerleLabs $PRL
Perle Labs Season 1 Snapshot: Locked In 🔒 This milestone showed the true scale of the community building sovereign, human-verified AI data infrastructure. Final points are now recorded and secured. You can check yours on the leaderboard: app.perle.xyz/ Thank you to everyone who showed up, completed tasks, and helped shape this transformative season. Details on contributor rewards and the next chapter for Perle are coming soon 🫡
1
7
129
My Liquid Trading battlestation 💧 📊 Charts on the main screen 📱 Market news on the tablet 💻 Trading on the laptop And two lucky stuffed companions are keeping the trades green. 🍀 @liquidtrading #liquidtrading
3
14
274
My entry for the first AI Prompt Battle for the community by @PerleLabs Theme: The Future of AI in Web3 Curious to see how everyone imagines the future of AI in Web3. Join the event: discord.com/events/140825578… Let’s cook 🔥 #perlelabs
1
10
92
gLiquid. The market only has two moods. 📈 Long Axel – “we’re so back.” 📉 Short Axel – “I told you it would dump.” Either way… Axel is trading. Are you Long or Short today?💧 @liquidtrading #axel #liquidtrading
5
16
153
If you passed the Perle Human CAPTCHA... Congrats. You’ve unlocked a special edition SBT 😎 This is your official onchain badge for proving you can tell what’s real from AI. If you submitted a valid claim, it’ll be making its way to your wallet shortly. Thanks to everyone who joined in on the fun!
390
177
64
1,176
64,203
New quest from @PerleLabs: Clinical Documentation Error Detection This one is the most challenging quest so far for me. It took me 30 minutes to complete it. It's easy to identify the error, but the tricky part is highlighting the exact error sentence. For example, in the image the documentation error is the anticoagulation plan, which means the text span containing the error: "bridge with enoxaparin 1mg/kg BID" but the correct answer is "Hold warfarin and bridge with enoxaparin 1mg/kg BID". My advice is to try to highlight multiple phrases in 1 sentence instead of multiple sentences. #perlelabs
New quest live: Clinical Documentation Error Detection 🩺 Each clinical note contains one safety-critical documentation error hidden somewhere inside. What you'll do: 1. Read the note 2. Highlight the exact text where the error appears 3. Select only the minimum span needed Think you can spot the errors? 👀 Get started right now: app.perle.xyz/
1
7
131
Axel in his natural habitat. Snack, Boba Tea and Chart. Locked in! #liquidtrading #axel @liquidtrading @m4lkxx
4
11
125
Perle Labs: The Compounding Intelligence Machine @PerleLabs changes the math of AI training data by turning human expertise into a compounding data engine where quality and it even increases with every cycle. The 4-Step Flywheel of Verifiable Truth Verified Expert Annotation: High-fidelity data begins with human specialists who provide the nuanced context machines often miss. Human-Verified Model Input: These annotations are used to refine model inputs, ensuring the foundation of the AI is built on verified accuracy rather than synthetic noise. On-Chain Reputation Scoring: Every contribution is recorded on the blockchain. This creates a permanent, transparent ledger of a contributor's accuracy and reliability. -> Note: The original text mentions "Nn-chain," but the architectural diagram confirms this is On-chain reputation tracking. Reputation-Weighted Task Allocation: The system automatically funnels the most complex and specialized domain work to those with the highest verified reputations. The Result: Quality that Compounds As contributors build their "on-chain resumes," they unlock higher-tier opportunities. This creates a virtuous cycle: better experts lead to better data, which leads to smarter models, which then require even more sophisticated expert feedback. #perlelabs
Perle Labs turns human expertise into a compounding data engine. Every annotation feeds a reputation-weighted system designed to improve data quality over time: → Verified expert annotation → Human-verified model input → Nn-chain reputation scoring → Reputation-weighted task allocation 🔁 repeat As contributors build verified reputations, they earn access to more complex, specialized domain work. The result: a system where data quality improves every cycle.
1
5
115
Axel and the boys opened a 50x position on @liquidtrading... but the market hasn’t moved an inch. Now we just wait for the chart to move. #axel #liquidtrading
4
12
90
Axel of Wall Street! “Sell me this drop.” The market never sleeps. Neither does Axel. Trade whatever wherever whenever💧💸 #axel #liquidtrading @liquidtrading
6
12
138
Quick reminder for the @PerleLabs community 👀 Perle Labs is hosting Perle Open Mic: Community Recognition Night this Thursday. 🗓 March 5 ⏰ 4:30 PM UTC They’ll be: • introducing new mods • recognizing contributors stepping into Navigator roles • sharing the latest updates across Perle Always cool to see projects actually spotlight their community builders. If you're active in the ecosystem, this is a good one to join. See you in the stage! #perlelabs
3
6
65
Predict anything on Liquid. Now powered by @Polymarket.
58
31
15
294
33,352
300K followers already. 500K is next! @PerleLabs just celebrated 200K at the start of February… and now we’re heading into March at 300K. That kind of growth doesn’t happen for no reason. People are starting to realize something simple: AI isn’t just about bigger models. It’s about better data. Human-verified. Auditable. Sovereign. That vision is resonating. Big respect to all the contributors building this from the ground up. Lowkey bullish on where this goes next. #perlelabs
300K followers. One shared vision. We celebrated 200K followers at the start of February. Just weeks later, we’re heading into March at 300K! Grateful for the contributors and supporters shaping Perle Labs together. AI that's human-verified. Auditable. Sovereign 🤝
5
59
One of the most important and underrated angles of USDX is that it transforms a stablecoin from a passive store of value into an active economic coordination layer. Traditional stablecoins like USDC or USDT are economically inert. They maintain price stability, but the underlying yield generated from their reserve assets (such as U.S. Treasury bills) is captured entirely by the issuer. Users get stability, but they don’t participate in the system's monetary yield. USDX introduces a fundamentally different model. Because USDX is backed by short-duration Treasury bills and cash equivalents, it generates real, external yield at the reserve layer. But instead of isolating that yield at the issuer level, USDX integrates it directly into the protocol’s economic architecture through a distribution mechanism aligned with network participation. #nexus #USDX #nexusUSDX
L1 + exchange + native stablecoin = new economic trifecta
1
5
52
Adversarial data poisoning is emerging as one of the most critical threats to modern AI systems. Instead of attacking the model directly, attackers target the training data, the foundation of how AI learns. Research like the OpenClaw attack shows how malicious actors injected harmful skills into AI agent ecosystems, allowing credential theft and system compromise through trusted pipelines. This type of attack is difficult to detect because poisoned data can appear normal while altering the model’s internal behavior. The risk becomes even greater in multi-agent systems. Studies on Single-agent Utility Shifting Attacks (SUSA) demonstrate that poisoning just one agent can shift the equilibrium of an entire system, causing cascading behavioral changes across all connected agents. This reveals a fundamental truth: AI security depends on data integrity. If training data is compromised, model outputs become unreliable by design. This is where @PerleLabs plays a critical role. Perle Labs enables the creation of verified ground truth data through human annotation, ensuring datasets are accurate, traceable, and resistant to manipulation. Each verified annotation strengthens the reliability of training datasets and reduces the risk of hidden poisoning. As AI systems increasingly power financial infrastructure, autonomous agents, and real-world decision-making, securing the data layer becomes essential. Trustworthy AI begins with trustworthy data. #perlelabs
2
6
70