Deepika Bantu (@dpkaa) profile picture
Joined April 2009
Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)Photo by Deepika Bantu (@dpkaa)
#SovrenContest I've run nodes on various Testnets for a while now. What I love about Sovren: the work is storage, delivery and compute for apps people already use. Now i am applying to run a Sovren node. @sovrentech #SovrenOperator
7
Deepika Bantu retweeted
Holder count up. Liquidity flat. Toddy trusts the second number more. New wallets are cheap. Depth is not. #Cryptocurrency
1
1
21
Deepika Bantu retweeted
J.P. Morgan Asset Management reportedly included Bitcoin in its 2027 Long-Term Capital Market Assumptions for the first time. Its projected annualized returns over the next 10–15 years for ₿ Bitcoin: 10% But here’s the part Toddy won’t ignore: Bitcoin volatility: ~55% annualized. 👀 Higher expected return. Much bigger swings. That’s a long-term assumption—not a short-term target, a guarantee, or a smooth ride. Would you rather hold the higher-volatility asset—or take the quieter ride? #TODDY #Bitcoin #Crypto
1
1
27
Naye Challenger spots ab available hain. Jaise-jaise hum aur advanced tasks introduce kar rahe hain, waise-waise hum un contributors ke liye access badha rahe hain jo lagataar high-quality data provide kar rahe hain. Challenger badges automatically award kiye jaate hain, jo aapke contribution ki quality aur current data collection ki zarooraton par depend karte hain. Iske liye manually apply karne ki koi process nahi hai. Apna Challenger badge mila hai ya nahi, yeh check karne ke liye apna portfolio check karein.
New Challenger spots are now available. As we introduce more advanced tasks, we’re expanding access for contributors who consistently deliver high-quality data. Challenger badges are awarded automatically based on contribution quality and current data collection needs. There’s no manual application process. Check your portfolio to see if you’ve received a Challenger badge.
27
15
8
54
1,090
Humne Axis Robotics ko 0 se Base par top robotics project kaise banaya 🟦? • Base Batches 003 ke finalists mein jagah banayi • NS mein Base Residency ka hissa bane • @donofbase ki taraf se Founder/Product content mila • @Nibel_eth ke saath 1-on-1 marketing strategy call mila • @baseapac ke saath INDO, HK aur PH mein community offline events co-host kiye • Base India aur Latam ke saath livestreams ke through product introduction kiya • @brian_armstrong ke saath Founder interview hua Aur in sab ke liye humne ek bhi paisa pay nahi kiya Yeh sab tabhi possible hota hai jab aap @buildonbase par build karte ho
31
13
48
517
Identity verification for the Axis Community Sale mein multiple precise steps hote hain. Participants ko verification process efficiently complete karne mein help karne ke liye, yeh guide step-by-step overview provide karti hai ki verification workflow kaise operate karta hai aur ise successfully kaise complete kiya ja sakta hai. 🧵
Identity verification for the Axis Community Sale involves multiple precise steps. To help participants navigate the process efficiently, this guide provides a step-by-step overview of how the verification workflow operates and how to complete it successfully. 🧵
18
11
1
33
821
Deepika Bantu retweeted
.@axisrobotics continues to demonstrate solid growth within the Base ecosystem. The project launched its main product on Base in March 2026 after large-scale tests. Anyone with a browser can teleoperate simulated robots and produce training data. No special hardware is required; physics run locally via WebAssembly. Every accepted trajectory receives a unique Data ID recorded on-chain on Base. That provenance trail links task, contributor, and dataset for auditability. The model is Train-to-Earn: quality-scored work earns points and future rewards. By late July 2026 it reported 94,000-plus global contributors and 2.1 million trajectories. On-chain records already exceeded 2.1 million across more than 1,600 published tasks. By early September the Hub was cited as a top-3 dApp on Base with 4.7 million-plus trajectories. More than 200,000 distributed contributors were active across thirteen embodiments. A $12 million seed round closed in July, led by Hack VC with Pi Network among backers. Base official accounts and Base APAC have repeatedly highlighted the project. Partnerships now span Booster Robotics, Unitree, AgiBot, Lotus Cars, and Geely Auto. Simulation data has already improved real-robot priors with far fewer physical demos. An open-source Franka-arm simulation dataset was released as one of the largest of its kind. Ego-centric mobile collection and human-gated post-training lines are scaling in parallel. Alliance tasks also feed BitRobot and other on-chain robotics networks. Low fees and EVM compatibility on Base make high-volume signing and incentives practical. The flywheel is simple: more users, more diverse data, better policies, more demand. Physical AI is no longer confined to closed labs; it is being built in public on Base. Axis is turning everyday human skill into verifiable, compounding robotic intelligence. s.kaito.ai/qTXv5bg
🤝 Paid partnership
.@quipnetwork makes the block itself useful work instead of a discarded hash. 
Optimization jobs that clients already pay for become the proof that extends the chain. 
Spare CPUs, GPUs, ASICs, and QPUs stop sitting idle and start producing answers plus consensus. 
Blockchain ceases to be a puzzle mill and becomes infrastructure for real computation. 
The same economy wraps existing chains in post-quantum accounts and atomic swaps. 
Crypto does not wait for a quantum break; it ships the defense on the chains people already use. 
Heterogeneous hardware and quantum-resistant keys prepare the stack for the machines that follow. 
Quip turns the ledger into a factory that computes today and survives tomorrow.
5
3
1
7
263
🥝🤖 place the kiwi into the microwave Me: “YES! Mission accomplished!” 😎🚀 Axis: “Nope. Slots finished.” 💀😭 One kiwi today. A smarter robot tomorrow. 🤖🥝 @axisrobotics @axisroboticsin_ Join Axis Robotics 👇 hub.axisrobotics.ai/login?in…
🤖 Made with AI
3
13
149
Me: “I’ll just do one quick Axis task.” 🤖 Axis: Pick up the tomato. 🍅 Me 30 minutes later: WHERE IS THE TOMATO?! 😭😂 @axisrobotics @axisroboticsin_ hub.axisrobotics.ai/login?in…
🤖 Made with AI
3
7
47
Humein apne research blog mein feature karne ke liye @cicada_mm ka dil se shukriya Puri research piece zaroor padhein jahan humne data volume ko quality compromise kiye bina scale karne robot embodiments ko prioritize karne aur foundation model scale ka asal mein kya matlab hota hai is par apni soch share ki hai nitter.cf/cicada_mm/status/20933…
Grateful to @cicada_mm for featuring us in their research blog. Dive into the full piece for our thinking on scaling data volume without sacrificing quality, how we prioritize robot embodiments, and what foundation-model scale actually means.
23
12
1
46
827
Axis Weekly Pichhle hafte humne browser policy server aur physics stack ke beech bache hue replay aur runtime gaps ko close kiya aur saath hi apne task generation engine (TaskGen) ki coverage ko scale kiya Ab articulated asset library generation pipeline ka hissa ban chuki hai, poora RoboCasa scene grid online hai, aur training aur distillation ke liye long-term aur short-term horizon tasks ka ek zyada clear split tayyar hai Key updates: • Replay & runtime: Policy aur human-trajectory replay ab frontend path ke saath match karte hain, aur web runtime ne full replay fidelity achieve kar li hai. Scenes ab zyada tezi se load hote hain, aur idle hone par simulation automatically pause ho jata hai. • Articulated assets & RoboCasa: Ab TaskGen mein 27 articulated object families ki library shamil hai, jisme har family ke 4 variants hain. Saare RoboCasa scenes ab 50×50 layout aur style grid ke roop mein online hain. • Horizon splits & next DAgger step: Long-term aur short-term horizon tasks ka ek zyada clean split tayyar hai, LIBERO Pro consistently high success rates dikha raha hai, aur agla round un correction segments ko filter karne par focus karega jo closed-loop behavior mein changes late hain. Neeche details dekhein 👇
Axis Weekly Last week, we closed the remaining replay and runtime gaps between the browser, policy server, and physics stack — then scaled the coverage of our task generation engine (TaskGen). An articulated asset library is now in the generation path, the full RoboCasa scene grid is online, and a cleaner long- versus short-horizon task split is ready for training and distillation. Key updates: • Replay & runtime: Policy and human-trajectory replay now match the frontend path, with web runtime reaching full replay fidelity. Scene loads faster, and simulation pauses automatically when idle. • Articulated assets & RoboCasa: TaskGen now includes a library of 27 articulated object families (4 variants each). All RoboCasa scenes are online as a 50×50 layout and style grid. • Horizon splits & next DAgger step: A cleaner long- vs. short-horizon split is ready, LIBERO Pro shows consistently high success rates, and the next round focuses on filtering correction segments that change closed-loop behavior. Details below 👇
29
10
2
69
1,806
🤖 New to Axis Robotics? Robot arm rotations can feel confusing at first. Here’s the easiest way I remember them: Save this for your next task. 👀🤖 What was the hardest control for you to learn? @axisrobotics @axisroboticsin_ Join fast👇 hub.axisrobotics.ai/login?in…
🤖 Made with AI
3
25
Today’s Axis challenge: Goal: Put tomato in blender 🍅 Problem: Find the tomato first. Status: Robot confused. Human confused. 😂 At this point we're both being trained. @axisrobotics
🤖 Made with AI
2
5
62
Axis Robotics has officially ruined my brain 😂 I don't pick up cups anymore. I calculate the grasp trajectory. MOVE → ROTATE → GRIP 🤖☕ Send help. @axisrobotics
🤖 Made with AI
2
4
50
CLAUDE + YouTube = $$$$ No degree. No camera. No editing skills. Even a 15-year-old can start. FULL blueprint FREE for the next 24 hours. Want the blueprint? 1. Like & RT 2. Comment “ AIMoney ” 3. Follow @zakiraicoder for DM. It's a money printer Don’t miss.
55
33
70
1,950
Deepika Bantu retweeted
🚨 66.1M Views on 1 Video 💰 Earned $8,570 from it ASMR Glass Slicing is the new Goldmine. Made in seconds using Veo 3 + right prompts. Want all the prompts I used? RT + Comment "Prompts" Follow, So I’ll DM everything you need
36
28
46
2,242
Deepika Bantu retweeted
Python is difficult to learn, but not anymore! Introducing Complete Python Training Course, All formulas, Tips & Tricks , shortcuts Absolutely Free It helps you to save 80% time and money To get it, just: 👉Follow @AnishJaitwar (so that I can DM) 👉 Like and Retweet 👉Comment "Send"
68
41
96
3,000
Deepika Bantu retweeted
Hiring! I'm looking for a new set of social media grinders to join my team. Salary will be $30 per hour No experience required, just eagerness to learn. Interested? Repost and Comment below👇 Make sure to follow me so we can dm you.
677
277
7
543
57,511