S.U (@SU__Crypto) profile picture
Joined March 2022
Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)Photo by S.U (@SU__Crypto)
πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ πŸ’΅ 250,000,000 $USDC (250,057,250 USD) minted at USDC Treasury whale-alert.io/tx/solana/4iH…
12
12
124
57,623
A robot completing a task makes for a great demo. Getting different robots, sensors and people to work together in the same constantly changing environment? That's a much bigger challenge. They need a common understanding of the space around them, where things are, where others are, and how the environment changes. That's why shared spatial context matters. It's the difference between isolated machines doing individual jobs and a connected system that can coordinate work across the physical world. The opportunity isn't just smarter robots. It's infrastructure that helps an entire ecosystem of machines understand the same world and work together. That's how Physical AI moves beyond impressive demos and toward real-world scale. - base:0xf9569cfb8fd265e91aa478d86ae8c78b8af55df4 🦾
Shared spatial context is what turns a single robot demo into a real production system. This is how human - robot teams actually scale. $Auki
2
10
410
S.U retweeted
Robots & Beers are the best events in the HK robotics community. Don't miss the next one on Nov 12, two weeks from today: luma.com/d8y1fsrc
11
54
1,383
S.U retweeted
Signed over 1000 locations already, with one new store coming online every day.
Nils @broodsugar talks about how @Auki will shift retail industry.
7
27
4
128
4,148
Nils @broodsugar talks about how @Auki will shift retail industry.
1
11
1
33
4,363
S.U retweeted
One new store per day is coming online with @Auki and @CactusXR right now, and it's accelerating.
14
29
3
130
4,298
S.U retweeted
Preview: no-code building of spatial data pipelines β€” drag and drop or ask your agent to do it. Example: Send video from robot A to node B which blurs people, then to node C running CV, then drop it in robot D's map.
1
17
2
60
2,756
S.U retweeted
Bring your own model (Qwen running on Jetson Thor in this case) to your domain, give it the spatial harness, and boom. Any model can leverage Cactus data, spatio-semantic or any other data in your domain, and even robots or other devices.
5
23
2
84
4,001
A quiet word about Bitcoin.
582
437
158
5,139
519,252
S.U retweeted
β–ͺ️ $AUKI β–ͺ️ Retail β†’ Agriculture 🚨 @Auki currently has thousands of active pilots within the retail sector… From small to large stores, all benefiting from their spatial intelligence tech suite Now they’re targeting an entirely new vertical: outdoor agriculture To achieve this they’ve partnered with Budbreak: Specialists in autonomous vineyard disease detection using AI & robots These robots can move through crops, inspecting the plants and grapes to: β†’ Detect early disease β†’ Monitor vine health β†’ Estimate crop yields But here’s where it gets interesting… β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” Physical Farms β†’ Digital Domains β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” Using Auki’s spatial computing tech… Data captured from Budbreak’s robots can be turned into gaussian splatting (As seen in the clip below) The result? A 3D reconstruction map of the entire vineyard… These maps can be viewed from a phone! Farmers can instantly see: 🟒 Areas producing Healthy vines πŸ”΄ Areas showing disease or stress 🟠 Exact locations that need attention β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” Fix problems before they spread β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” Farmers using this tech can now walk straight to the problem spots That means: ⚑ Faster intervention 🌱 Healthier crops 🦠 Less yield loss from disease β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” This is 'the real world web' in action β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” Robots scanning crops Spatial AI mapping farmland AR guiding farmers in the field Physical agriculture + digital intelligence All working together β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€” We’ve seen @Auki rollout their tech suite in major retail spaces successfully I think they’re about to do it again But this time we’re going outdoors
Scaling up our gaussian splatting experiments with a 100-meter domain. Data from Budbreak who's building perception robots for vineyards. πŸ‡
5
9
1
48
2,152
Everyone is talking about the robots, but I think people are overlooking what makes them useful in the real world. A robot needs to know where it is, what’s around it and where it needs to go. base:0xf9569cfb8fd265e91aa478d86ae8c78b8af55df4 handles that well. Cactus is already doing work in retail, and the team is pushing further into spatial mapping and robotics. Still early but my conviction in @Auki is Rock Solid. Cant shake the destiny of these Legends. πŸ”₯
Robotics has taken over BIG time and the one project I am so bullish on is @Auki Auki isn't just talking about the future of Physical AI. Cactus is already a retail AI copilot, with Auki reporting a paid software business. The team has demonstrated robots using maps built by phones to localize in a changed environment, and Cactus directing a robot to investigate a shelf problem. These may look like separate developments, but I see one bigger picture taking shape. Retail intelligence. Spatial mapping. Robot navigation. Shared context between devices. Less friction when deploying machines into new environments. The pieces are starting to connect. My thesis is that the next phase of Physical AI won't be won by better robots alone. It'll also need infrastructure that helps those robots understand the world they're operating in. That's the layer Auki is building with the Real World Web. And base:0xf9569cfb8fd265e91aa478d86ae8c78b8af55df4 has a potential role beyond speculation: the protocol is designed around burning tokens to access network services and staking tokens to participate as an infrastructure provider. Here's the important distinction, though. Auki's token model connects the vision to real utility. Network services use credits purchased by burning AUKI, while infrastructure providers stake AUKI to participate. As the Real World Web expands, the protocol has a direct mechanism to connect network activity with token demand and supply reduction. And the opportunity goes far beyond retail robots. The same spatial infrastructure can serve phones, smart glasses, connected devices and autonomous machines operating in the physical world. That's why I see base:0xf9569cfb8fd265e91aa478d86ae8c78b8af55df4 as more than a robotics play. Auki is building the spatial intelligence layer for a world where AI doesn't just think, but understands its surroundings and acts within them. I'm backing the vision of spatial intelligence becoming foundational infrastructure for Physical AI, with AUKI at the core of the network's utility and participation model. The robots are the visible part. The infrastructure connecting them to the real world is the bigger story.
6
7
35
891
S.U retweeted
Scaling up our gaussian splatting experiments with a 100-meter domain. Data from Budbreak who's building perception robots for vineyards. πŸ‡
4
34
3
111
7,659
S.U retweeted
Wifi dead zones are common in large spaces, and can be pretty bad for robots especially if they're being tele-operated. Step 1 of fixing that is autonomous signal surveying and mapping, demoed here by our in-house robot Scentinel.
3
25
88
2,389
🚨 🚨 🚨 🚨 928 $BTC (75,436,661 USD) transferred from #OKEX to unknown wallet whale-alert.io/tx/bitcoin/3b…
6
7
1
66
77,776
S.U retweeted
Do you think retailers would spend 0.5% of their revenue on physical AI β€”Β the most powerful tool they've ever had? That's already $125 billion. It's a big pie, and we've got a head start on cutting out a slice.
How big can physical AI in retail get? Nils from @Auki shares the numbers here, from signed locations to 2027 deployments: β–ͺ️ Auki has already signed over 1,000 locations for @CactusXR β–ͺ️ The project plans to deploy over 1,000 robot store managers in 2027, connected to Cactus and delivered to existing customers β–ͺ️ Retail could spend over $125 billion on physical AI within 5 years This is what a clear plan looks like: a product and signed locations today, robot deployments planned for 2027 and over $125 billion in retail spending expected within 5 yearsπŸ‘€ imho the future looks bright 🦾
1
22
87
3,540
S.U retweeted
Going from demo store to live store means introducing people and other obstacles. Obviously "don't run into the shopper" is best learned in SIM, so we're working on that too with simulated pedestrians in a reconstruction of our office.
1
19
2
92
3,660
S.U retweeted
FYI - Last time $BTC had positive 3d EMAs crosses was in 2023, price was circa 27.5k. When on a ride to 73k over the year. Currently BTC Bulls defending 81.4k on 3d - chart in thread below. 1/2
5
4
11
775
S.U retweeted
Smart homes and spaces on steroids: give your physical AI harness to any LLM and you can query/prompt it. Of course, your robots plug right in, out of the box.
6
20
4
96
4,386
S.U retweeted
Nils recently gave a keynote at the Humanoid Hub conference on our core thesis of the real world web and why we're making environments smarter while everyone else is only focused on the robots. For yesterday's community update he went through it again. πŸ‘‡
1
22
1
75
3,618
S.U retweeted
$BTC If this is all the pullback we're getting that's incredible Nice lil bounce of honestly a meh support And all that amidst geopolitical turmoil and a broad expectation of a Black Monday Good stuff, long may it last
109
149
8
2,515
116,796