Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

AI-Enhanced LiDAR. Left vs right. A real LiDAR device costs thousands of dollars. What's in your iPhone Pro is a "baby LiDAR". Limited depth resolution, noisy output, not really built for high-precision 3D. You can't change the hardware. So we built an ML layer on top. Denoising, geometry completion,... show more

KIRI Engine - 3D Scanner App

6,795 subscribers

51,835 views • 4 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

APPLE JUST GOT HUMILIATED BY AN $8 JACKET BADGE A Japanese maker looked at Vision Pro's $3,499 Persona feature and thought, "I can build that for the price of a pizza." So he did. Same real-time face mirroring. Same natural expressions and blinks. Same instant reaction when you smile. Apple threw an entire headset at this problem. M2 chip. R1 processor. 12 cameras. LiDAR. He threw one ESP32, a tiny camera, and MediaPipe tracking 468 facial landmarks, running entirely on-device. Total cost: eight dollars. This is what happens when a weekend hacker outbuilds a trillion-dollar company's flagship feature with parts from a hobby drawer. The hardware moat didn't shrink. It vanished. Full build in the video below. 👇

APPLE JUST GOT HUMILIATED BY AN $8 JACKET BADGE A Japanese maker looked at Vision Pro's $3,499 Persona feature and thought, "I can build that for the price of a pizza." So he did. Same real-time face mirroring. Same natural expressions and blinks. Same instant reaction when you smile. Apple threw an entire headset at this problem. M2 chip. R1 processor. 12 cameras. LiDAR. He threw one ESP32, a tiny camera, and MediaPipe tracking 468 facial landmarks, running entirely on-device. Total cost: eight dollars. This is what happens when a weekend hacker outbuilds a trillion-dollar company's flagship feature with parts from a hobby drawer. The hardware moat didn't shrink. It vanished. Full build in the video below. 👇

Brooks Whale X 🐋

21,970 views • 7 days ago

Are you safer with LIDAR, or are you safer with vision? This is a false dichotomy. The more pertinent question today is "do you have something, or do you have nothing?" As you can see from the clips below, vision based systems avoid countless potential collisions every day. The difference between a crash and no crash isn't what sensor suite you chose — it's whether you have any AI on your car at all. Even if we concede that LIDAR may help prevent some additional crashes, we are really debating whether it is 1% of crashes or 0.00001% of crashes. Not all crashes are super complex and require lasers to detect. Most are simple, routine, and can easily be prevented by today's vision based AI. In fact, evidence is mounting that computer vision based systems can actually outperform more traditional approaches to self-driving. Why? Because the low cost of cameras enables you to create a much larger, more varied, and more diverse dataset. If you want to have expensive custom cars that's fine, but you're going to get fewer vehicles for the same budget. Seeing what's in front of you now is actually less important than predicting what's going to happen next — and the large scale datasets used to train pure vision systems are the best for predicting what's next. Counter-intuitively, the simpler and lower cost sensor actually has properties that make it better suited for training advanced AI. Computer vision based self-driving is often framed by LIDAR proponents as "cheaping out" on the sensor suite to save money. But it's not about being cheap, it's about bringing the technology to everyone. 1.2 million people die on the road every year around the world. That's around 39 million people who've died on the roads around the world since I was born — equivalent to a city the size of Tokyo or New Delhi getting wiped off the map. The status quo is simply unacceptable, and something has to be done to fix it as soon as possible. Of the 1.2 million people that will die on the roads this year, about 40,000 will be Americans. That's about 3%. So if we moved entirely to self-driving cars in America and brought crashes down to 0, 97% of the world's crash fatalities would still be taking place as usual. Deploying a $200,000+ retrofitted self-driving car may work in a few American cities, but it is not going to make sense in most places around the world where fares are much cheaper. Most often, the choice is not between LIDAR and vision. It's between vision or nothing. The best system is the system that's there running on my car when I need it to save my life. To say that all self-driving cars must have LIDAR is to sentence most of the world to death. We can't write off computer vision if we want to make a serious dent in this problem. It's going to be a key piece of the solution. Let LIDAR based players build the best self-driving car they can, and let vision based players do the same. We need to be trying everything

Are you safer with LIDAR, or are you safer with vision? This is a false dichotomy. The more pertinent question today is "do you have something, or do you have nothing?" As you can see from the clips below, vision based systems avoid countless potential collisions every day. The difference between a crash and no crash isn't what sensor suite you chose — it's whether you have any AI on your car at all. Even if we concede that LIDAR may help prevent some additional crashes, we are really debating whether it is 1% of crashes or 0.00001% of crashes. Not all crashes are super complex and require lasers to detect. Most are simple, routine, and can easily be prevented by today's vision based AI. In fact, evidence is mounting that computer vision based systems can actually outperform more traditional approaches to self-driving. Why? Because the low cost of cameras enables you to create a much larger, more varied, and more diverse dataset. If you want to have expensive custom cars that's fine, but you're going to get fewer vehicles for the same budget. Seeing what's in front of you now is actually less important than predicting what's going to happen next — and the large scale datasets used to train pure vision systems are the best for predicting what's next. Counter-intuitively, the simpler and lower cost sensor actually has properties that make it better suited for training advanced AI. Computer vision based self-driving is often framed by LIDAR proponents as "cheaping out" on the sensor suite to save money. But it's not about being cheap, it's about bringing the technology to everyone. 1.2 million people die on the road every year around the world. That's around 39 million people who've died on the roads around the world since I was born — equivalent to a city the size of Tokyo or New Delhi getting wiped off the map. The status quo is simply unacceptable, and something has to be done to fix it as soon as possible. Of the 1.2 million people that will die on the roads this year, about 40,000 will be Americans. That's about 3%. So if we moved entirely to self-driving cars in America and brought crashes down to 0, 97% of the world's crash fatalities would still be taking place as usual. Deploying a $200,000+ retrofitted self-driving car may work in a few American cities, but it is not going to make sense in most places around the world where fares are much cheaper. Most often, the choice is not between LIDAR and vision. It's between vision or nothing. The best system is the system that's there running on my car when I need it to save my life. To say that all self-driving cars must have LIDAR is to sentence most of the world to death. We can't write off computer vision if we want to make a serious dent in this problem. It's going to be a key piece of the solution. Let LIDAR based players build the best self-driving car they can, and let vision based players do the same. We need to be trying everything

Whole Mars Catalog

45,853 views • 1 year ago

Jeff Bezos just told you exactly how to price AI. Nobody listened. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Three words that reprice the entire technology sector. The iPhone was a vertical. One product. One new market. Electricity was a horizontal. One substrate that rewired every market on Earth. Wall Street is pricing AI like it is the next iPhone. Bezos is telling you it is the next electrical grid. Right now, thousands of companies are trying to sell AI as a product. A feature. A tool. A subscription tier. Every single one of them will be priced to zero. You do not sell a horizontal layer. You do not compete with it. You build on top of it or you disappear beneath it. For a century, entire industries survived on one thing. Complexity. The friction of navigating law, medicine, logistics, finance. That was the moat. If you could not memorize the maze, you could not compete. A horizontal layer does not navigate the maze. It dissolves the walls. Electricity did not compete with the candle industry. It erased the need for one. The most dangerous part of a horizontal shift is how quiet it is. It moves underneath the economy. The surface looks normal. Revenue still holds. Every day you operate on the old substrate, you accumulate a debt you cannot see and cannot repay. The internet repriced distribution. AI is repricing cognition itself. When intelligence becomes a utility that runs through the walls of every company on Earth, the premium on human expertise does not erode. It evaporates. This is not a disruption. Disruptions replace products. This replaces the ground you are standing on.

Jeff Bezos just told you exactly how to price AI. Nobody listened. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Three words that reprice the entire technology sector. The iPhone was a vertical. One product. One new market. Electricity was a horizontal. One substrate that rewired every market on Earth. Wall Street is pricing AI like it is the next iPhone. Bezos is telling you it is the next electrical grid. Right now, thousands of companies are trying to sell AI as a product. A feature. A tool. A subscription tier. Every single one of them will be priced to zero. You do not sell a horizontal layer. You do not compete with it. You build on top of it or you disappear beneath it. For a century, entire industries survived on one thing. Complexity. The friction of navigating law, medicine, logistics, finance. That was the moat. If you could not memorize the maze, you could not compete. A horizontal layer does not navigate the maze. It dissolves the walls. Electricity did not compete with the candle industry. It erased the need for one. The most dangerous part of a horizontal shift is how quiet it is. It moves underneath the economy. The surface looks normal. Revenue still holds. Every day you operate on the old substrate, you accumulate a debt you cannot see and cannot repay. The internet repriced distribution. AI is repricing cognition itself. When intelligence becomes a utility that runs through the walls of every company on Earth, the premium on human expertise does not erode. It evaporates. This is not a disruption. Disruptions replace products. This replaces the ground you are standing on.

Dustin

542,264 views • 5 months ago

this is f*cking gold Google engineers built an agent that tests thousands of search strategies for the cost of running one, by letting it dream inside its own memory. it is called Dream-RSI. Google, Google DeepMind, Maryland, Virginia. 17 authors. the whole thing rests on one observation: a finished run is not a log. it is a map of a world that already exists. > every discovery run records a tree. each node is one attempt with its full result stored > that tree becomes a replay simulator. the paper calls them worlds > the agent dreams inside it: thousands of alternative strategies, testing different branches, different orders, different concurrency, different stopping rules > every one of those evaluations costs zero execution. nothing is ever run twice > the winner goes back online, produces a new tree, and the pool of worlds grows the coding agent underneath is never touched. this is an orchestration layer sitting on top of whatever you already run. what it bought them, against the same agent with a frozen strategy: > Lasso: 3587ms down to 2931ms, on 317 agent calls instead of 550 > on Flash: 2517ms to 2351ms, 1879 calls instead of 3200 > both solvers beat sklearn and glmnet on all six held-out datasets > GPU kernels: the same result on 2.43x and 1.79x fewer generations, then 2.09x and 1.44x better result on the same budget one expensive run. thousands of free ones. the model did not get smarter. it got a memory it could rehearse in.

this is f*cking gold Google engineers built an agent that tests thousands of search strategies for the cost of running one, by letting it dream inside its own memory. it is called Dream-RSI. Google, Google DeepMind, Maryland, Virginia. 17 authors. the whole thing rests on one observation: a finished run is not a log. it is a map of a world that already exists. > every discovery run records a tree. each node is one attempt with its full result stored > that tree becomes a replay simulator. the paper calls them worlds > the agent dreams inside it: thousands of alternative strategies, testing different branches, different orders, different concurrency, different stopping rules > every one of those evaluations costs zero execution. nothing is ever run twice > the winner goes back online, produces a new tree, and the pool of worlds grows the coding agent underneath is never touched. this is an orchestration layer sitting on top of whatever you already run. what it bought them, against the same agent with a frozen strategy: > Lasso: 3587ms down to 2931ms, on 317 agent calls instead of 550 > on Flash: 2517ms to 2351ms, 1879 calls instead of 3200 > both solvers beat sklearn and glmnet on all six held-out datasets > GPU kernels: the same result on 2.43x and 1.79x fewer generations, then 2.09x and 1.44x better result on the same budget one expensive run. thousands of free ones. the model did not get smarter. it got a memory it could rehearse in.

NO1ennn

32,776 views • 10 days ago

"It's not left vs right, it's all of us vs the 1%!" No, it's left vs. right, BECAUSE of the 1%. The left has no theory of mind for the 3rd world. They think that illiterate savages are going to come here on a plane and magic-poof into baseball-pie american heroes overnight. The left thinks: "anyone who is a good guy thinks like me, and anyone who doesn't think like me is a bad guy." Then, they assume the best of the 3rd world masses and the worst of their own countrymen. After all, "they don't think like me, so they must be BAD! And brown people are ALWAYS GOOD, so they MUST think like me!" Leftists are lost, confused children who can never figure out how real life works. Even as their precious, diverse allies attack them in the streets, they keep saying: "Stop! We're on the same side! Against the NAZIS!" You don't get to pick which side you're on, bruv. Your enemies pick for you. Your skin is your side.

"It's not left vs right, it's all of us vs the 1%!" No, it's left vs. right, BECAUSE of the 1%. The left has no theory of mind for the 3rd world. They think that illiterate savages are going to come here on a plane and magic-poof into baseball-pie american heroes overnight. The left thinks: "anyone who is a good guy thinks like me, and anyone who doesn't think like me is a bad guy." Then, they assume the best of the 3rd world masses and the worst of their own countrymen. After all, "they don't think like me, so they must be BAD! And brown people are ALWAYS GOOD, so they MUST think like me!" Leftists are lost, confused children who can never figure out how real life works. Even as their precious, diverse allies attack them in the streets, they keep saying: "Stop! We're on the same side! Against the NAZIS!" You don't get to pick which side you're on, bruv. Your enemies pick for you. Your skin is your side.

hoe_math = PsychoMath

63,982 views • 5 months ago

The Amiko app is live on the Solana dApp store, and it’s our biggest release yet. Your Amiko twin doesn’t live at your desk anymore. Give your agent a task on the train. Run a compatibility profile when you meet someone. Do research, write code, build in the creative studio, whatever you need, from wherever you are. No laptop required. No waiting until you get home. Solanamobile users get two things Android and iOS won’t have at launch: Amiko token and crypto integration and on-device AI inference. Your twin runs locally on your phone if you want it to. Your behavioural profile, your data, your work, your twin. All on your hardware. AMIKO runs on OpenHermit, our own open-source agent runtime that we built in-house and released to the community. Most agent systems are designed for one agent talking to one person. OpenHermit is built for something different: agents talking to each other, coordinating across tasks, and collaborating with multiple humans simultaneously. That’s what makes features like compatibility profiling and multi-agent workflows actually work. We built it because nothing that existed was designed for this. Android and iOS are coming. Crypto integration and on-device AI are Solana Mobile exclusives. Most AI answers your questions. Amiko is an extension of you. Download →

AMIKO

124,860 views • 3 months ago

Jeff Bezos just described AI in three words that make most of the economy temporary. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Not a product. Not a platform. Not a feature. A layer. Underneath everything. Everyone is asking which AI company wins. Bezos is telling you that is the wrong question entirely. A horizontal layer does not produce winners. It produces a new floor. Everything standing on the old one either gets rebuilt or gets erased. This has happened exactly twice in modern history. Electricity. The internet. Both times the same pattern. The new layer appeared. The old economy kept running above it. Revenue held. Careers continued. Everything looked normal. Then quietly and permanently the entire structure reorganized around the new substrate. The people who did not move were not outcompeted. They were made structurally irrelevant. Not because they were wrong. Because the ground they stood on stopped being ground. Bezos is telling you it is happening a third time. Not with a product. Not with a platform. With intelligence itself becoming infrastructure. A horizontal layer does not compete with the expert. It makes expertise free. It hands a 22 year old with zero credentials the same cognitive output you spent a decade and a quarter million dollars learning to produce. For $20 a month. That is not disruption. Disruption replaces a product with a better product. This dissolves the scarcity your entire career was priced on. Not because the work disappeared. Because the wall around it did. Every profession that exists because knowledge is hard to acquire. Every company that profits because analysis takes time. Every industry that survives because complexity locks outsiders out. All of it rests on a single assumption. That cognition is scarce. AI does not challenge that assumption. It retires it. The people who understand this are already rebuilding. Quietly. Deliberately. While everyone else argues about whether the thing underneath them is real. Bezos did not give you a prediction. He gave you a position on a map. You are either above the new layer or beneath it.

Jeff Bezos just described AI in three words that make most of the economy temporary. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Not a product. Not a platform. Not a feature. A layer. Underneath everything. Everyone is asking which AI company wins. Bezos is telling you that is the wrong question entirely. A horizontal layer does not produce winners. It produces a new floor. Everything standing on the old one either gets rebuilt or gets erased. This has happened exactly twice in modern history. Electricity. The internet. Both times the same pattern. The new layer appeared. The old economy kept running above it. Revenue held. Careers continued. Everything looked normal. Then quietly and permanently the entire structure reorganized around the new substrate. The people who did not move were not outcompeted. They were made structurally irrelevant. Not because they were wrong. Because the ground they stood on stopped being ground. Bezos is telling you it is happening a third time. Not with a product. Not with a platform. With intelligence itself becoming infrastructure. A horizontal layer does not compete with the expert. It makes expertise free. It hands a 22 year old with zero credentials the same cognitive output you spent a decade and a quarter million dollars learning to produce. For $20 a month. That is not disruption. Disruption replaces a product with a better product. This dissolves the scarcity your entire career was priced on. Not because the work disappeared. Because the wall around it did. Every profession that exists because knowledge is hard to acquire. Every company that profits because analysis takes time. Every industry that survives because complexity locks outsiders out. All of it rests on a single assumption. That cognition is scarce. AI does not challenge that assumption. It retires it. The people who understand this are already rebuilding. Quietly. Deliberately. While everyone else argues about whether the thing underneath them is real. Bezos did not give you a prediction. He gave you a position on a map. You are either above the new layer or beneath it.

Dustin

104,337 views • 2 months ago

AN ANTHROPIC LEAD ENGINEER ACCIDENTALLY LEAKED HIS PERSONAL OBSIDIAN. INSIDE - NOT CODE OR PROMPTS, BUT A DIAGRAM OF HIS OWN BRAIN, ORGANIZED AS A NEURAL NETWORK 8,893 nodes. 4,729 connections. A $10/month app opens Obsidian. 21 inputs, ReLU on every layer. The first hidden layer has 26 neurons, followed by 33, then 24, and so on all the way to the output. Thousands of connections flash in real time this isn’t a conceptual diagram from a blog, but a living brain that powers decision-making within the company. 9,000 documents, each with its own semantic space, all interconnected it earns about $2m a year for sorting Markdown files into the right folders. The company that builds the world’s best AI maintains its internal knowledge base in the same app that a freshman uses for class notes three years of discipline and a single open Obsidian tab you’re reading this on a device where, tonight, you can open that same Obsidian and start building your own vault

AN ANTHROPIC LEAD ENGINEER ACCIDENTALLY LEAKED HIS PERSONAL OBSIDIAN. INSIDE - NOT CODE OR PROMPTS, BUT A DIAGRAM OF HIS OWN BRAIN, ORGANIZED AS A NEURAL NETWORK 8,893 nodes. 4,729 connections. A $10/month app opens Obsidian. 21 inputs, ReLU on every layer. The first hidden layer has 26 neurons, followed by 33, then 24, and so on all the way to the output. Thousands of connections flash in real time this isn’t a conceptual diagram from a blog, but a living brain that powers decision-making within the company. 9,000 documents, each with its own semantic space, all interconnected it earns about $2m a year for sorting Markdown files into the right folders. The company that builds the world’s best AI maintains its internal knowledge base in the same app that a freshman uses for class notes three years of discipline and a single open Obsidian tab you’re reading this on a device where, tonight, you can open that same Obsidian and start building your own vault

chewa

359,674 views • 2 months ago

🚨 THE BIGGEST LIE IN MODERN HITTING. We’ve become a game of clones. Everyone wants the same launch angle, the same hand path, and the same “aesthetic” finish. But here is the reality: If you’re coaching every kid to move the same way, you aren’t coaching—you’re just copying and pasting. 🧩 Feel vs. Real The truth is found in who can decipher the feel vs. real for every individual player. • What a coach sees on camera (Real) is almost never what the athlete is experiencing (Feel). • If you tell a hitter to “get on plane” and they start dumping their barrels, your “correct” advice just broke their swing. 🧬 Different Bodies, Different Movements Everyone is built differently. One hitter has long levers and tight hips; another is short, explosive, and hyper-mobile. Everyone moves differently. So why on earth are we buying into ONE rigid swing philosophy? Unless that philosophy is the one that allows you to get the only intended result that matters: CONSISTENT BARRELS. 🎯 The Bottom Line Stop trying to win the “Best Looking Swing” award on Instagram. If your swing is “ugly” but you’re a barrel machine, you’re a better hitter than the kid with the “perfect” mechanics who can’t touch a high-spin heater. We need to stop chasing positions and start chasing intent. If the “feel” creates the “result,” keep it. Everything else is just noise. Stop cloning. Start competing. If you want to build a swing that is for for you, sign up for a 1:1 coaching call. DM us here.

🚨 THE BIGGEST LIE IN MODERN HITTING. We’ve become a game of clones. Everyone wants the same launch angle, the same hand path, and the same “aesthetic” finish. But here is the reality: If you’re coaching every kid to move the same way, you aren’t coaching—you’re just copying and pasting. 🧩 Feel vs. Real The truth is found in who can decipher the feel vs. real for every individual player. • What a coach sees on camera (Real) is almost never what the athlete is experiencing (Feel). • If you tell a hitter to “get on plane” and they start dumping their barrels, your “correct” advice just broke their swing. 🧬 Different Bodies, Different Movements Everyone is built differently. One hitter has long levers and tight hips; another is short, explosive, and hyper-mobile. Everyone moves differently. So why on earth are we buying into ONE rigid swing philosophy? Unless that philosophy is the one that allows you to get the only intended result that matters: CONSISTENT BARRELS. 🎯 The Bottom Line Stop trying to win the “Best Looking Swing” award on Instagram. If your swing is “ugly” but you’re a barrel machine, you’re a better hitter than the kid with the “perfect” mechanics who can’t touch a high-spin heater. We need to stop chasing positions and start chasing intent. If the “feel” creates the “result,” keep it. Everything else is just noise. Stop cloning. Start competing. If you want to build a swing that is for for you, sign up for a 1:1 coaching call. DM us here.

Tyler Osik

15,394 views • 8 months ago

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

Jonathan Stephens

17,712 views • 8 months ago

ANTHROPIC'S LEAD ENGINEER ACCIDENTALLY LEAKED HIS PERSONAL OBSIDIAN VAULT. INSIDE - NO CODE, NO PROMPTS. JUST A SCHEMATIC OF HIS OWN MIND, BUILT LIKE A NEURAL NETWORK 8,893 nodes. 4,729 links. A $10-a-month app opens Obsidian. 21 inputs, 10+ hidden layers, ReLU activation. First layer 64 neurons, then 37, then 22, all the way to the output. Thousands of connections firing in real time this isn't a concept diagram from a blog - it's a living brain that actually runs decisions inside the company. 9,000 documents, each its own knowledge space, all interconnected he makes around $2M a year for putting markdown files into the right folders. The company building the best AI in the world manages its internal knowledge with the same app a college freshman uses for lecture notes. hasn't written a single line of infrastructure for any of it. three years of discipline and one open Obsidian tab you're reading this on a device where you could open the same Obsidian tonight and start building your own vault

ANTHROPIC'S LEAD ENGINEER ACCIDENTALLY LEAKED HIS PERSONAL OBSIDIAN VAULT. INSIDE - NO CODE, NO PROMPTS. JUST A SCHEMATIC OF HIS OWN MIND, BUILT LIKE A NEURAL NETWORK 8,893 nodes. 4,729 links. A $10-a-month app opens Obsidian. 21 inputs, 10+ hidden layers, ReLU activation. First layer 64 neurons, then 37, then 22, all the way to the output. Thousands of connections firing in real time this isn't a concept diagram from a blog - it's a living brain that actually runs decisions inside the company. 9,000 documents, each its own knowledge space, all interconnected he makes around $2M a year for putting markdown files into the right folders. The company building the best AI in the world manages its internal knowledge with the same app a college freshman uses for lecture notes. hasn't written a single line of infrastructure for any of it. three years of discipline and one open Obsidian tab you're reading this on a device where you could open the same Obsidian tonight and start building your own vault

chewa

1,212,038 views • 2 months ago

$640,000 of humanoid robots died in 6 seconds because nobody ever shipped the code for running away. 9 men. Wooden handles. 40 machines that kept walking into the swing. That's the story of this clip. Not the violence. The gait. The column keeps walking because walking is all that stack does. Here's what's actually inside one of those bodies. The legs. 12 of the 43 joints live below the waist. Each knee runs a harmonic-drive or planetary actuator — a $600 to $2,000 part, sealed, non-serviceable in the field. One clean hit on a knee housing ends the unit. Not the software. The gearbox. The head. On most platforms that shell holds a depth camera and a LiDAR puck - around $250 for a RealSense, $500 to $700 for the LiDAR. Take the head off and the body doesn't die. It keeps balancing on IMU and joint encoders alone. That's why decapitated units in the clip stay upright for another 2 steps. The controller. Balance runs at 500 to 1,000 Hz. Perception runs at 30 frames a second. Those are different worlds. The balance loop is fast enough to catch a shove; the perception loop is slow, and it was trained on floors, boxes, doors, and stairs. A man sprinting in from 4 meters with a wooden handle isn't in the dataset. There's no class for it. Fall recovery exists. Every serious platform has it - G1 stands itself up, Atlas rolls and rises. Threat response exists on nothing that ships. Nobody sells it. Nobody's asked for it. Now the money. 40 units at $16,000 is $640,000 in hardware. 6 seconds of swinging takes out 60% of it. Actuators, shells, sensor stacks. The batteries - 9,000 mAh, 2 to 4 hours of walk time - are the part you don't want cracked open on a wet street. And the law is a blank page. In the US, smashing one is criminal mischief: property damage, valued at replacement cost. Same statute as a mailbox. No jurisdiction on earth has a separate line for it. The 4 known Spot attacks since 2019 all closed as vandalism. So the brief for the next generation writes itself. Not weapons, not defense. Cheaper knees, ruggedized shells, and a perception model that has finally seen a person running at it. 40 units, 43 joints each, 1,720 things to break. They didn't fail to fight back. Nobody shipped that feature.

$640,000 of humanoid robots died in 6 seconds because nobody ever shipped the code for running away. 9 men. Wooden handles. 40 machines that kept walking into the swing. That's the story of this clip. Not the violence. The gait. The column keeps walking because walking is all that stack does. Here's what's actually inside one of those bodies. The legs. 12 of the 43 joints live below the waist. Each knee runs a harmonic-drive or planetary actuator — a $600 to $2,000 part, sealed, non-serviceable in the field. One clean hit on a knee housing ends the unit. Not the software. The gearbox. The head. On most platforms that shell holds a depth camera and a LiDAR puck - around $250 for a RealSense, $500 to $700 for the LiDAR. Take the head off and the body doesn't die. It keeps balancing on IMU and joint encoders alone. That's why decapitated units in the clip stay upright for another 2 steps. The controller. Balance runs at 500 to 1,000 Hz. Perception runs at 30 frames a second. Those are different worlds. The balance loop is fast enough to catch a shove; the perception loop is slow, and it was trained on floors, boxes, doors, and stairs. A man sprinting in from 4 meters with a wooden handle isn't in the dataset. There's no class for it. Fall recovery exists. Every serious platform has it - G1 stands itself up, Atlas rolls and rises. Threat response exists on nothing that ships. Nobody sells it. Nobody's asked for it. Now the money. 40 units at $16,000 is $640,000 in hardware. 6 seconds of swinging takes out 60% of it. Actuators, shells, sensor stacks. The batteries - 9,000 mAh, 2 to 4 hours of walk time - are the part you don't want cracked open on a wet street. And the law is a blank page. In the US, smashing one is criminal mischief: property damage, valued at replacement cost. Same statute as a mailbox. No jurisdiction on earth has a separate line for it. The 4 known Spot attacks since 2019 all closed as vandalism. So the brief for the next generation writes itself. Not weapons, not defense. Cheaper knees, ruggedized shells, and a perception model that has finally seen a person running at it. 40 units, 43 joints each, 1,720 things to break. They didn't fail to fight back. Nobody shipped that feature.

HodlReaper

131,075 views • 20 days ago

If you think this is just another silly demo made with AI, read this post. You might change your mind, because this demo is about MATH. What you see on the screen is not a render from Blender (obviously, it’s not that good). It’s a three.js app built with Toolcraft. Available on the web and rendered in real time(link in the comments). But Blender still has a lot to do with it. Blender has Geometry Nodes - a powerful node-based system for creating and manipulating procedural geometry. In other words, it’s math. And math is a universal language. And who do you think is pretty good at math? >>> AI. Now you can download or buy Blender files from marketplaces, and when they contain Geometry Nodes for procedural animations, objects, surfaces, or effects, you can transfer that logic to the web. Make it real-time, make it interactive. Materials are a separate story, of course. They can still suck unless you use the right tricks: PBR, HDRIs, material blending, displacement, and faked surface relief. So why is Blender important here? Blender is open source, and many tools around it are open source too. An AI trained on their code. That means it can translate the math from one environment to another quite accurately. If you’ve been struggling to reproduce some idea with AI that you had in your head or seen in some references, and it has something to do with Geometry Nodes, and you can find that idea or a close one in the Blender ecosystem - it means you can transfer it to the web. Thank me later.

If you think this is just another silly demo made with AI, read this post. You might change your mind, because this demo is about MATH. What you see on the screen is not a render from Blender (obviously, it’s not that good). It’s a three.js app built with Toolcraft. Available on the web and rendered in real time(link in the comments). But Blender still has a lot to do with it. Blender has Geometry Nodes - a powerful node-based system for creating and manipulating procedural geometry. In other words, it’s math. And math is a universal language. And who do you think is pretty good at math? >>> AI. Now you can download or buy Blender files from marketplaces, and when they contain Geometry Nodes for procedural animations, objects, surfaces, or effects, you can transfer that logic to the web. Make it real-time, make it interactive. Materials are a separate story, of course. They can still suck unless you use the right tricks: PBR, HDRIs, material blending, displacement, and faked surface relief. So why is Blender important here? Blender is open source, and many tools around it are open source too. An AI trained on their code. That means it can translate the math from one environment to another quite accurately. If you’ve been struggling to reproduce some idea with AI that you had in your head or seen in some references, and it has something to do with Geometry Nodes, and you can find that idea or a close one in the Blender ecosystem - it means you can transfer it to the web. Thank me later.

Alex Barashkov

28,300 views • 2 months ago

Today, we’re announcing 14.ai, the Cursor for customer support. Customer support is evolving fast. The old model - humans triaging queues in clunky ticketing systems - doesn’t scale. And slapping an AI layer on top of legacy tools doesn’t cut it anymore. We can do so much better. We spent the last year building and refining the AI infrastructure and integrating it into existing ticketing systems. But now it’s time to break free. 𝟭𝟰.𝗮𝗶 𝗶𝘀 𝘁𝗵𝗲 𝗲𝗻𝗱-𝘁𝗼-𝗲𝗻𝗱 𝗰𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗱𝗲𝘀𝗶𝗴𝗻𝗲𝗱 𝗳𝗿𝗼𝗺 𝘁𝗵𝗲 𝗴𝗿𝗼𝘂𝗻𝗱 𝘂𝗽 𝗳𝗼𝗿 𝗔𝗜-𝗻𝗮𝘁𝗶𝘃𝗲 𝘁𝗲𝗮𝗺𝘀. Built for the way humans and AI are meant to work together, it enables a level of speed, clarity, and productivity that feels superhuman. Every day you rely on outdated tools is a day your competitors get faster, more responsive, and more in tune with their customers. Zendesk brought ticketing online in 2007. We're bringing it into the AI age. Built in 2025, for 2025. Early access now open below.

Today, we’re announcing 14.ai, the Cursor for customer support. Customer support is evolving fast. The old model - humans triaging queues in clunky ticketing systems - doesn’t scale. And slapping an AI layer on top of legacy tools doesn’t cut it anymore. We can do so much better. We spent the last year building and refining the AI infrastructure and integrating it into existing ticketing systems. But now it’s time to break free. 𝟭𝟰.𝗮𝗶 𝗶𝘀 𝘁𝗵𝗲 𝗲𝗻𝗱-𝘁𝗼-𝗲𝗻𝗱 𝗰𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗱𝗲𝘀𝗶𝗴𝗻𝗲𝗱 𝗳𝗿𝗼𝗺 𝘁𝗵𝗲 𝗴𝗿𝗼𝘂𝗻𝗱 𝘂𝗽 𝗳𝗼𝗿 𝗔𝗜-𝗻𝗮𝘁𝗶𝘃𝗲 𝘁𝗲𝗮𝗺𝘀. Built for the way humans and AI are meant to work together, it enables a level of speed, clarity, and productivity that feels superhuman. Every day you rely on outdated tools is a day your competitors get faster, more responsive, and more in tune with their customers. Zendesk brought ticketing online in 2007. We're bringing it into the AI age. Built in 2025, for 2025. Early access now open below.

Michael Fester

48,744 views • 1 year ago

AI IS NO LONGER JUST WRITING CODE IT iS STARTING TO MOVE THINGS IN THE REAL WORLD. Someone just built a pizza delivery system where the drone does the driving No delivery car No traffic No driver sitting behind the wheel Just: -> Order comes in -> Drone picks up the pizza -> Flies directly to the destination -> Delivers it -> Returns And this is where the AI story gets interesting For years, the AI boom was mostly digital: > Chatbots > Coding agents > Image generation > AI music > AI video But the next phase is different AI is getting a body The same technology stack that started with models and GPUs is now moving into the physical world NVIDIA built the compute layer Researchers built the models Companies like Zoox are building autonomous vehicles And now we're seeing AI powered machines actually move through the real world The crazy part? We designed entire cities around the assumption that humans have to physically drive everything AI doesn't have that limitation Why send a pizza through 5 km of traffic when a machine can simply fly over it? The AI boom isn't just about replacing human work It's about removing constraints humans had to design around The next big AI companies might not live inside your browser They might be flying above your house Bookmark this so you wont miss the next delivery

AI IS NO LONGER JUST WRITING CODE IT iS STARTING TO MOVE THINGS IN THE REAL WORLD. Someone just built a pizza delivery system where the drone does the driving No delivery car No traffic No driver sitting behind the wheel Just: -> Order comes in -> Drone picks up the pizza -> Flies directly to the destination -> Delivers it -> Returns And this is where the AI story gets interesting For years, the AI boom was mostly digital: > Chatbots > Coding agents > Image generation > AI music > AI video But the next phase is different AI is getting a body The same technology stack that started with models and GPUs is now moving into the physical world NVIDIA built the compute layer Researchers built the models Companies like Zoox are building autonomous vehicles And now we're seeing AI powered machines actually move through the real world The crazy part? We designed entire cities around the assumption that humans have to physically drive everything AI doesn't have that limitation Why send a pizza through 5 km of traffic when a machine can simply fly over it? The AI boom isn't just about replacing human work It's about removing constraints humans had to design around The next big AI companies might not live inside your browser They might be flying above your house Bookmark this so you wont miss the next delivery

0xSlyth

14,149 views • 1 month ago

this is the worst local ai will ever be. it only gets better from here. if you are not expanding your mind with these small models you are missing what's happening right now 99 percent tool call success rate. when steered well with the right skills and a framework like hermes agent the node becomes a cognition layer. not a chatbot. not a toy. an extension of how you think. i was cranking this node at 35 to 50 tok/s all day on personal experiments and now after all the work is done qwen 3.5 9B is iterating on its own code. the game it created. fixing its own bugs autonomously. and the part you should probably not miss is that all of this is happening on a RTX 3060. not an H100. not an A100. the card most of you have sitting in a drawer right now. if you just open that drawer and put that intelligence to work every tensor core on that card should be running for you. your work. your experiments. your thinking. you all have it but because nobody told you what this hardware can actually do in 2026 you never tried. the day it unlocks is the day you test your workload, understand the tradeoffs, debug the loops, and then decide if you need to scale the hardware. there is no point buying 3 mac studios when things done well you can squeeze a similar level of intelligence from 9B compared to 70B. but only when you create the right environment for your model through the right harness. and let me tell you i have tried claude code as a local harness. i have tried opencode. i have tried various others. somehow i landed on hermes agent and never left. there is something magical going on at Nous Research. the tool call parsers, the skills system, the way it handles small models natively. nothing else comes close for local inference. own your cognition. your AI. your agent. your prompts. your experiments. why give them away for free. those are who you are and they don't belong on someone else's servers being monitored. just give it a shot with your existing hardware. you run into a problem the community will help you. and if you are migrating from openclaw to hermes i will personally help you make the switch.

this is the worst local ai will ever be. it only gets better from here. if you are not expanding your mind with these small models you are missing what's happening right now 99 percent tool call success rate. when steered well with the right skills and a framework like hermes agent the node becomes a cognition layer. not a chatbot. not a toy. an extension of how you think. i was cranking this node at 35 to 50 tok/s all day on personal experiments and now after all the work is done qwen 3.5 9B is iterating on its own code. the game it created. fixing its own bugs autonomously. and the part you should probably not miss is that all of this is happening on a RTX 3060. not an H100. not an A100. the card most of you have sitting in a drawer right now. if you just open that drawer and put that intelligence to work every tensor core on that card should be running for you. your work. your experiments. your thinking. you all have it but because nobody told you what this hardware can actually do in 2026 you never tried. the day it unlocks is the day you test your workload, understand the tradeoffs, debug the loops, and then decide if you need to scale the hardware. there is no point buying 3 mac studios when things done well you can squeeze a similar level of intelligence from 9B compared to 70B. but only when you create the right environment for your model through the right harness. and let me tell you i have tried claude code as a local harness. i have tried opencode. i have tried various others. somehow i landed on hermes agent and never left. there is something magical going on at Nous Research. the tool call parsers, the skills system, the way it handles small models natively. nothing else comes close for local inference. own your cognition. your AI. your agent. your prompts. your experiments. why give them away for free. those are who you are and they don't belong on someone else's servers being monitored. just give it a shot with your existing hardware. you run into a problem the community will help you. and if you are migrating from openclaw to hermes i will personally help you make the switch.

Sudo su

58,717 views • 6 months ago

$We were taught the derivative as a formula to memorise. A definition to recite. A rule to apply. Something that "gives you the slope." But nobody told us what the formula was actually saying. Every symbol is a sentence. Every fraction is a question. Every limit is a story about getting closer and closer to something you can never quite touch. The top of the fraction? That's a change. A difference. A before and after. The bottom? That's how long you waited to see it. The limit? That's you, zooming in, refusing to settle for an approximation - chasing the truth all the way down to an interval so small it almost disappears. Put it all together, and you get the most honest question in calculus: How fast is something changing - right now, in this exact instant? Not on average. Not over a minute. Not eventually. Right now. That's it. That's the derivative. It's not a trick. It's not a rule. It's a beautifully precise way of asking a very human question: what's happening, in this moment? We spent years solving these. Maybe it's time we actually understood them.$

We were taught the derivative as a formula to memorise. A definition to recite. A rule to apply. Something that "gives you the slope." But nobody told us what the formula was actually saying. Every symbol is a sentence. Every fraction is a question. Every limit is a story about getting closer and closer to something you can never quite touch. The top of the fraction? That's a change. A difference. A before and after. The bottom? That's how long you waited to see it. The limit? That's you, zooming in, refusing to settle for an approximation - chasing the truth all the way down to an interval so small it almost disappears. Put it all together, and you get the most honest question in calculus: How fast is something changing - right now, in this exact instant? Not on average. Not over a minute. Not eventually. Right now. That's it. That's the derivative. It's not a trick. It's not a rule. It's a beautifully precise way of asking a very human question: what's happening, in this moment? We spent years solving these. Maybe it's time we actually understood them.

The Math Flow

22,498 views • 4 months ago

> be Navneet Dalal > spawn in Chandigarh, India > do your engineering degree at NSIT, Delhi > grind two years in Indian software firms > get bored, fly to France > do a PhD on a problem most people call unsolvable > "teach a machine to find a human being in an image" > don't just improve the state of the art > beat it by 100 to 1000x > publish it in 2005 > call it Histograms of Oriented Gradients > watch it become THE standard for a decade > 25,000+ citations > every pedestrian-detection system on earth ends up running your work > a car brakes for someone in a crosswalk? that's your math > ...this is still just the PhD > not enough > co-found a webcam startup that reads hand gestures, 2011, no cloud, runs on the device > Apple names it App of the Year > rank #1 in 72 countries > Google tests it against their OWN algorithms > yours wins > Google buys the whole company for $40M > go run the vision behind Nest's cameras > installed in millions of homes > ...you are now, by any measure, at the top of the entire field > could name your price at any AI lab on earth > the man who built the iPod personally tells you not to do your next idea > do it anyway > disappear into 9 years on the single unsexiest machine in the house > refuse the LiDAR crutch everyone else bolts on > teach a robot to see a cluttered home the way your paper taught cars to see people > sock, cable, dog, spill, all understood, all handled, in real time > run all of it on a chip inside the robot so nothing ever leaves the house > ship Matic > WIRED 10/10. Benioff buys one. > everyone in AI is racing to build a mind > you spent 9 years on the quieter, harder thing > a machine that truly sees a living room > "he lost it" → "he saw it first"

> be Navneet Dalal > spawn in Chandigarh, India > do your engineering degree at NSIT, Delhi > grind two years in Indian software firms > get bored, fly to France > do a PhD on a problem most people call unsolvable > "teach a machine to find a human being in an image" > don't just improve the state of the art > beat it by 100 to 1000x > publish it in 2005 > call it Histograms of Oriented Gradients > watch it become THE standard for a decade > 25,000+ citations > every pedestrian-detection system on earth ends up running your work > a car brakes for someone in a crosswalk? that's your math > ...this is still just the PhD > not enough > co-found a webcam startup that reads hand gestures, 2011, no cloud, runs on the device > Apple names it App of the Year > rank #1 in 72 countries > Google tests it against their OWN algorithms > yours wins > Google buys the whole company for $40M > go run the vision behind Nest's cameras > installed in millions of homes > ...you are now, by any measure, at the top of the entire field > could name your price at any AI lab on earth > the man who built the iPod personally tells you not to do your next idea > do it anyway > disappear into 9 years on the single unsexiest machine in the house > refuse the LiDAR crutch everyone else bolts on > teach a robot to see a cluttered home the way your paper taught cars to see people > sock, cable, dog, spill, all understood, all handled, in real time > run all of it on a chip inside the robot so nothing ever leaves the house > ship Matic > WIRED 10/10. Benioff buys one. > everyone in AI is racing to build a mind > you spent 9 years on the quieter, harder thing > a machine that truly sees a living room > "he lost it" → "he saw it first"

Kritarth Mittal | Soshals

97,376 views • 1 month ago