ASHIK (@Ashik_80) header
ASHIK (@Ashik_80) profile picture

Content Creator/Writer R-Mod @axisrobotics

Joined May 2025
Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)Photo by ASHIK (@Ashik_80)
Pinned Tweet
Most Comprehensive Robotics Data Projects 1. @axisrobotics We can say Axis is the most proven and successful robotics project in Web3 at this moment. They are building a compounding data engine for physical AI. They have also partnered with multiple Web2 robotics companies. 2. @BitRobotNetwork BitRobot is trying to build a data network layer for robotics. They have already onboarded some robotics projects including Axis Robotics. 3. @openroboto OpenRoboto operates a robotics model-improvement network on Bittensor Subnet 80. Their main focus is model training and competition, as well as decentralized robotics-data collection. 4. @PrismaXai PrismaX is a robotics teleoperation and data marketplace. 5. @InvLambda Inverted Lambda is building an economic layer around physical-AI contributions. 6. @ZenO4AI ZenO focuses on first-person, real-world data for physical AI. 7. @4Dlabs_Official 4D Labs is focused on 3D/4D spatial intelligence. It’s not a final list. This ranking is based on the current market scenario and some projects may do far better than others. It’s just a research result based on current data. And @SurfAIHQ helped me a lot in finding the exact rank.
13
5
1
43
1,907
Guys... don’t bully those who farm roles on @RialoHQ discord Not every DYOR can give you a big shot, just like not every trade makes money. It’s not the contributors fault. They’re just trying their best to earn something. They’re already broken, no need to push them harder. Let’s STAY Positive and STAY Happy 💛
11
2
26
885
big week...! so its coming... $MM or what?
6
7
213
Every zkTLS project hits the same wall “Users won’t install a Chrome extension” That friction has killed more privacy apps than bad cryptography ever did @primus_labs just removed it Their Lightweight Page SDK runs the full zkTLS flow from a normal web page >No extension >No install prompt >No permission dialogs. Integrate it into a webpage, and attestation happens client-side without changing the familiar user experience. Built for real consumer flows: - prove income without uploading statements - show bank balance for DeFi credit - verify follower count or creator revenue on-chain One integration No extension dependency They still keep the heavier options: Network JS SDK for extension based, high-security user flows Network Core SDK for backend/institutional pipelines Different tools for different threat models zkTLS was never blocked by the crypto It was blocked by distribution A page level SDK that works like every other web library is the fastest path to real adoption
Replying to @primus_labs
2/ Primus’s AutoHoG automatically designs custom gates that take multiple inputs and produce multiple outputs, then folds mergeable subcircuits into a single bootstrapping. On 8 consumer RTX 5090s, one 64-bit confidential transfer now takes about 90 ms, with throughput above 210 per second. Everything runs in integer arithmetic, so nodes produce byte-identical results.
75
11
83
692
Everyone is calling for another BTC low this time... is it really going to happen? First, we have to accept that nobody can predict the future. But we can imagine it, if we have enough data. To imagine what BTC might do in the next 3 to 6 months, we need to analyze the past 3 to 6 months. So I analyzed the data to see whether the outlook is positive or negative and @SurfAIHQ helped me pull the real data on time. What I analyzed: - Crypto infrastructure and cybersecurity incidents - Geopolitical and political uncertainty - AI and semiconductor stress - BTC market structure - BTC ETF outflows - Liquidations and on-chain behavior - RSI, Fear & Greed, and market sentiment It’s easy to say “support is here, resistance is there.” But without proper research, why should anyone believe those words? BTC’s price doesn’t depend on a single factor, so I tried to look at all of them. 1. Crypto Infrastructure and Cybersecurity Incidents We all know about the recent Binance security concerns, where hackers reportedly took $6.3B in funds. This was the trigger for much of the bearish sentiment. Binance is one of the biggest and most reputed CEXs in the industry and some users even said “Binance is too big to fall.” But Binance wasn’t the only incident. We saw cybersecurity incidents happen one after another. TSMC, the world’s biggest semiconductor manufacturer, had a data leak and the hacker demanded 3 BTC for the claimed records. This also fueled the bearish sentiment. 2. Geopolitical and Political Uncertainty The US midterm results had a huge impact on the geopolitical picture, with the Democratic Party winning by a big margin. On top of that, uncertainty has recently increased everywhere: Russia, Europe, the Middle East, China. 3. AI and Semiconductor Stress This is the data that is forcing investors to rethink their positions. The tech incident wasn’t a single outage. It was a growing risk-off narrative around AI and semiconductors, with huge uncertainty around US-based AI companies. What we noticed in the past few days: - US AI-stock losses - Concerns about tech valuations - Semiconductor scrutiny - Nvidia leadership uncertainty - The upcoming Anthropic IPO All of this puts pressure on bullish sentiment in crypto. BTC is increasingly traded as a high-beta liquidity asset, and when investors reduce exposure to expensive tech stocks, they often reduce exposure to crypto too. 4. BTC Market Structure BTC’s one-year data shows a significant deterioration from late-2025 prices. It’s down 13.5% in 7 days and 20.7% in 30 days, a clear bearish trend over the last month. Two bullish signals to watch: - Reclaiming $73K–$76K would be the first sign of stabilization. - Reclaiming $80K–$84K would be more meaningful, since it would recover the 50-day average and the former breakdown supply. 5. BTC ETF Outflows Last year, ETFs played a vital role in the bullish sentiment. This year it’s the opposite. Bitcoin ETFs reportedly saw about $2.1B in November outflows, including a seven-day outflow streak. Market coverage also linked the decline to rate-hike expectations and inflation concerns. 6. Liquidations and On-Chain Behavior We recently saw a huge washout. Most over-leveraged long traders have either closed their positions or been liquidated. Open interest is around $41.5B, with roughly $99B in 24-hour futures volume. 7. RSI, Fear & Greed and Sentiment Daily RSI is around 21.9, which is technically oversold. The Fear & Greed Index sits at 11, Extreme Fear. This suggests we could see a relief rally soon, or one more leg down toward $55K–$58K first. Result: - If BTC can’t hold $62K–$64K, expect the $55K–$58K area. - If BTC reclaims $68K–$70K, market sentiment could start to shift. - Strong resistance sits at $87K–$92K. All of this holds only if we don’t see fresh uncertainty like in the past few months. Not financial advice. Do your own research.
3
2
8
155
Next Support for $BTC ? 63K or 58K or another new LOW ? 🙂
5
7
151
Axis Sim-Lab Training আসছে! Axis Bangladesh কমিউনিটির জন্য আয়োজন করা হচ্ছে একটি লাইভ ট্রেনিং সেশন। এই সেশনে আমরা শিখব কীভাবে Axis Hub-এ Simulation Tasks সম্পন্ন করতে হয়, কীভাবে রোবট অপারেট করতে হয় এবং সাধারণ ভুলগুলো এড়িয়ে আরও ভালোভাবে কাজ করা যায়। 🎙️ Host: @Ashik_80 ⏱️ Duration: 30 Minutes 📍 Platform: Discord সেশনে যা থাকছে: - Axis Hub-এর পরিচিতি ও ব্যবহার - Robot Operation এবং Live Practical Demonstration - অভিজ্ঞ Contributors-দের Tips & Tricks - Q&A এবং Community Feedback নতুন হোন বা অভিজ্ঞ Contributor, সবার জন্যই এই সেশনটি হতে যাচ্ছে শেখার এবং নিজেদের দক্ষতা বাড়ানোর দারুণ একটি সুযোগ। সবাইকে অংশগ্রহণের আমন্ত্রণ!
1
2
8
138
We’ve been using ODE Policy for visual policy distillation, with strong early results. Learning how actions evolve over time gives the policy a natural way to capture the continuity of robot motion. In the video, our Franka robot learns to thread a nut onto a bolt in simulation, reproducing the task from SGS. Next, we plan to use ODE Policy to distill a wider range of assembly skills into policies that act from camera images, from fitting gears and inserting pegs to joining connectors.
🚀 Nature does not move in discrete jumps. Why should robot policies? Super excited to release ODE Policy — perhaps our most exciting breakthrough in my lab this year! 🤖🔥 Think about how humans reach for a cup. Our movements evolve continuously, governed by underlying dynamics. Yet, most robot policies — from Diffusion and Flow Matching to VLAs — learn to predict explicit actions. What if robots learned how actions evolve, instead of predicting actions? 💡 That's ODE Policy for Embodied AI. We use a neural Ordinary Differential Equation (ODE) to learn continuous action dynamics, rather than directly generating actions. One simple idea. Five types of robots. Multiple learning paradigms. ✅ Single-arm manipulation ✅ Dual-arm manipulation ✅ Dexterous hands ✅ Humanoid whole-body control ✅ Aerial robots And it goes far beyond imitation learning: ⚡ VLA (π0.5): 96.0% vs. 75.5% success after one epoch. 🌍 4.5B World Action Model: 82.6% vs. 24.6% success, with ~4× faster action generation. 🏃 RL: More efficient humanoid whole-body motion learning. For years, we've focused on scaling robot data, models, and compute. But what if the next fundamental breakthrough comes from rethinking WHAT robots learn, rather than just HOW MUCH they learn? Huge congratulations to Jindou Jia and our amazing team! 👏 📄 Paper: arxiv.org/abs/2611.01729 🌐 Project: ntumars.github.io/ODE/ #ODEPolicy #EmbodiedAI #PhysicalAI #RobotLearning #WorldModels #VLA #MARSLab
24
20
8
166
11,281
What makes an alpha trader successful? I think a trader who can turn negative news into a money-making opportunity is a successful one. We all saw the #NVIDIA CEO headlines, right? But can you tell me who actually made money from this news? Anyone? I just found an interesting Solana trader... @rasmr_eth Amid the Jensen Huang/NVIDIA headlines, he spotted the meme narrative around $PUSSY (with the NVIDIA logo) at just ~$143K market cap. FomoLens shows the position closed with ~$23.57K in realized profit (+196.5%). I’m no one to decide who is successful and who isn’t, but I really love how @rasmr_eth turned negative news into profit. While others were searching for something more interesting, he was busy finding alpha. Not only that, he made over $1.1M in just 3 trades, though he also lost $578K. On @fomo 30-day ranking he sits at #15 with +$97.78K monthly and +$491.71K all-time PnL. His biggest win came from BNB Chain, but his biggest loss came from Solana. Tell me about yourself... did you make any money on the Jensen Huang/NVIDIA headlines? Research powered by @SurfAIHQ
11
2
13
249
How should you choose simulation demos for co-training when real robot data is scarce? Introducing TUCO: Curating Simulation Demonstrations for Sim-to-Real Robot Policy Co-Training, the first systematic study of data curation for sim-to-real robot policy co-training. With just 10 real demos and 600 curated sim demos per task, TUCO hits 85.8% real-robot success, versus 25.8% for the full 1,200 pool or a random 600, and sets SOTA across all three transfer settings. A collaboration with researchers from @Stanford @NTUsg @CarnegieMellon @FudanUniversity @UCBerkeley @sjtu1896. Project page: tuco-curation.github.io/ Details below ⬇️
40
27
9
171
10,934
It’s really sad to see that @Galxe is going to shut down their Quest platform. I have a lot of memories with @GalxeQuest. I started my crypto journey with a big win that came from @GalxeQuest. My big win was @spaceandtime. I received more than $500 just by doing some simple quests on Galxe. What days those were... But yeah, you have to improve yourself based on market demand. Best of luck... @Galxe team!
Galxe is consolidating its team and brand under Gravity to accelerate the development and adoption of Gravity L1. Over the past five-plus years, we’ve helped pioneer Web3 questing and given projects new ways to connect with users and build communities. We’ve served 36 million users and supported more than 7,700 partner projects through campaigns, credentials, and loyalty programs, helping people discover new projects, contribute to their ecosystems, and earn recognition and rewards. We’re bringing that experience and the relationships built along the way into our work on Gravity. As part of this consolidation, we will wind down Galxe Quest, Galxe ID, Galxe Passport, and related services. Gravity will continue operating as usual, with no changes to the G token as part of this transition. Starting today, Spaces can no longer create new campaigns. New Galxe+ and Galxe Score subscriptions and Galxe Passport minting are also no longer available. December 5, 2026, will be the last day to participate in campaigns. No campaigns will remain active after that date. On December 15, 2026, campaign reward claims will close, and the entire Galxe website will switch to read-only mode. Please claim any outstanding campaign rewards before December 15. User assets remain safe, including Smart Savings balances, staked G that has not yet been withdrawn, and G tokens held in wallets on Gravity Alpha Mainnet. These assets will follow a separate claim process and timeline. We will publish the details and instructions in a follow-up announcement on our official X. Thank you to the users, builders, and partners who helped shape Galxe. We look forward to continuing to build with you on Gravity.
7
11
263
kaito (@KaitoAI), Ink (@inkonchain) and many more their paid customers... To be honest their tech and partnership looks promising But at the end @yzilabs honest with u their reputation not that gd specially in recet times.. But I believe @primus_labs can do better...
.@primus_labs has 1M+ users & 10+ enterprises paying for its zkTLS infra, backed by @yzilabs the numbers so far: > $6.5M raised in a pre-seed + seed in Feb 2025 > 1,000,000+ users across the extension, Dev Hub & zkTLS apps > 10+ paid enterprise customers across stablecoins, AI, L2s & socials > $72M+ TVL verified through their proof of reserves across 4 live projects the GTM is B2B2C, they make money from API fees, gas that scales with volume & enterprise setup + management fees developers > apps > network usage > enterprises > more developers ok ✍️
4
1
1
9
190
Last day I was free and went out to hang out with my friends... One of my friends asked me... “Bro, last time u told me about @primus_labs how it works and its concept.” Me... “Yes... that was a pretty techy conversation.” My friend... “Ya... but I got a Marvel character that perfectly fits Primus.” Me... “ahhh... Which one?” My friend... “The Watcher.” Me... “Haha, u’re right bro... but Primus is more than that.” My friend... “How?” Me... “Think about the Watchers. They see everything, they record everything... but they never interfere. And if someone asks ‘is that true?’ all they can say is ‘trust me, I saw it.’” My friend... “Yeah... right” Me... “Right. That’s where Primus goes one step beyond. It doesn’t only watch... it helps u prove what it watched.” My friend... “Prove it how?” Me... “Say u log in to ur exchange. Primus checks that the data really comes from the real website. Then u can prove one fact, like ‘my balance is above this or that number... without showing the whole account. The website doesn’t even have to change anything.” My friend... “So no ‘trust me bro’... just proof.” Me... “Exactly. And FHE adds one more thing. Even while the data is being used, it can stay encrypted.” My friend... “So that means... a verifiable watcher?” Me... “Exactly, u’re right. A watcher that can prove what it saw and still keep ur private data private.” My friend smiled... “Okay, that’s actually a cool way to remember it.” Me... "ahhh yah ... " @primus_labs
last day when I posted about how Primus and zkTLS work together... My friend asked me... “Don’t u tell me that @primus_labs works with FHE also?” Me... “Yes, they use FHE.” My friend... “Then tell me, how do they use it?” Me... “FHE means Fully Homomorphic Encryption. Sounds a little technical, right...?” My friend... “Ahh, it sounds like something scientific.” Me... “Hahaha. The easy meaning is this: a computer can work on ur data while the data stays locked.” My friend... “Wait... isn’t encrypted data already safe?” Me... “Yes, but only while it’s sitting somewhere. To use it, a computer normally has to unlock it first. And at that moment, someone can see it.” My friend... “Oh... so that’s the weak point.” Me... “Exactly and FHE removes it. Think of an accountant. u give him a locked box full of bills. He adds them up without ever opening the box. Then he gives u back a locked box with the total inside. Only u have the key.” My friend... “So he did the work but never saw my data?” Me... “Yes. That’s FHE.” My friend... “Okay, but where does Primus use it?” Me... “Two places. First, the Confidential Vault. Ur amounts and position stay hidden from the public. Second, FHETransform, which brings FHE to smart contracts on EVM chains. The inputs, the contract state and the outputs all stay encrypted.” My friend... “Does the blockchain do all this?” Me... “No. We all know that FHE math is heavy... so the heavy work runs off-chain. If the chain did everything, it would be slow and costly.” My friend... “So zkTLS and FHE are the same thing?” Me... “not really. zkTLS proves that data is real. FHE keeps data private while it’s being used.” My friend smiled and said... “got u... tnx for this info” Me... “ Mention not...” @primus_labs
6
11
197
ASHIK retweeted
🚨 Warning for all NFT hunters! Please be careful with projects like @AxiomARB Today, My friend @Spaulhrx connected his wallet to mint nft of this project and immediately afterward, his wallet was compromised. Always DYOR before connecting your wallet to any unfamiliar project. And use burner Wallet Stay safe, protect your assets, and never blindly trust any project. ⚠️
🚨 X just updated its algorithm on November 5, bringing new changes for creators, especially small accounts. Likes now carry more weight, increasing from 0.66 to 0.708. Reading time and shares have also become more valuable. Meanwhile, simply opening a post or video engagement is worth less. Videos now need genuine watch time to generate value. Another important change: AI generated replies under large accounts may get flagged, meaning your replies might never reach the intended audience. The first two hours remain crucial for engagement. The new strategy is simple: Create original content that keeps people reading, encourages likes, and sparks genuine conversations. Quality engagement matters more than ever. Are you changing your posting strategy?
19
6
3
35
2,570
Finally, it’s over... @InvLambda S3 (GENESIS) has officially ended. So what’s next? Something NFT type? Or something rewards type? Ahh... why not both 👀
THIRD CONTACT: GENESIS IS OFFICIALLY CLOSED Inverted Lambda wouldn’t be here without our community. Every run and vision shared brought us across the finish line. While the web platform is still live, data has been snapshotted. Final scores, leaderboards, and reward distributions will be announced mid-next week. Something is on the way 👀.
6
14
231
last day when I posted about how Primus and zkTLS work together... My friend asked me... “Don’t u tell me that @primus_labs works with FHE also?” Me... “Yes, they use FHE.” My friend... “Then tell me, how do they use it?” Me... “FHE means Fully Homomorphic Encryption. Sounds a little technical, right...?” My friend... “Ahh, it sounds like something scientific.” Me... “Hahaha. The easy meaning is this: a computer can work on ur data while the data stays locked.” My friend... “Wait... isn’t encrypted data already safe?” Me... “Yes, but only while it’s sitting somewhere. To use it, a computer normally has to unlock it first. And at that moment, someone can see it.” My friend... “Oh... so that’s the weak point.” Me... “Exactly and FHE removes it. Think of an accountant. u give him a locked box full of bills. He adds them up without ever opening the box. Then he gives u back a locked box with the total inside. Only u have the key.” My friend... “So he did the work but never saw my data?” Me... “Yes. That’s FHE.” My friend... “Okay, but where does Primus use it?” Me... “Two places. First, the Confidential Vault. Ur amounts and position stay hidden from the public. Second, FHETransform, which brings FHE to smart contracts on EVM chains. The inputs, the contract state and the outputs all stay encrypted.” My friend... “Does the blockchain do all this?” Me... “No. We all know that FHE math is heavy... so the heavy work runs off-chain. If the chain did everything, it would be slow and costly.” My friend... “So zkTLS and FHE are the same thing?” Me... “not really. zkTLS proves that data is real. FHE keeps data private while it’s being used.” My friend smiled and said... “got u... tnx for this info” Me... “ Mention not...” @primus_labs
8
1
16
424
We're building a multi-layered robotics architecture, where the LLM drives high-level reasoning and skill harness. Here's a demo. The task: use the right arm to arrange three objects from the tray in a row along the X axis of the table, ordered by height from tallest to shortest. You can see the agent turn the task into continuous motion through skill harness, linking scene understanding, height-based ordering, and placement planning to physical execution. This is an exciting start.
This is already the present for SOTA robotics architecture
32
17
10
179
11,223