Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

Axis Product Update The Precision & Security Patch is here. - Hardened Anti-Cheat: Deployed backend trajectory replay verification and end-state DB checks. Bots and invalid runs are strictly filtered. - Wrist-Cam View: Added a direct camera feed from the robot's wrist for precise manipulation. - Custom Sensitivity: A new... show more

Axis Robotics

37,751 subscribers

19,134 görüntüleme • 5 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

At Axis, every trajectory submitted by our community undergoes a strict replay validation process. We run each submission through checker to verify whether the target task was successfully completed. To see how strict it is, check this demo (Task: Place The Toy Train On The Board Game Box). Real human data passes smoothly (Video 1). However, bots or manually altered data will fail (Video 2). Why? Faking numbers breaks the simulation's physics causality. Even tiny tweaks cause error accumulation, resulting in failed movements. This invalid data is automatically rejected. Because of this mechanism, data submitted via bots will ultimately fail our replay verification. Invalid data is strictly excluded from model training, and the task slot is reopened to the community to collect genuine, high-quality trajectories. Furthermore, we actively monitor for duplicated data. Trajectories that are identical lack the diversity required for robot learning and will not be credited by our scoring system. If we detect accounts submitting a massive volume of identical trajectories, all associated addresses will be permanently banned. For Axis, the quality and diversity of data are the only ways to solve the robotics generalization gap. They will always be our absolute top priorities.

At Axis, every trajectory submitted by our community undergoes a strict replay validation process. We run each submission through checker to verify whether the target task was successfully completed. To see how strict it is, check this demo (Task: Place The Toy Train On The Board Game Box). Real human data passes smoothly (Video 1). However, bots or manually altered data will fail (Video 2). Why? Faking numbers breaks the simulation's physics causality. Even tiny tweaks cause error accumulation, resulting in failed movements. This invalid data is automatically rejected. Because of this mechanism, data submitted via bots will ultimately fail our replay verification. Invalid data is strictly excluded from model training, and the task slot is reopened to the community to collect genuine, high-quality trajectories. Furthermore, we actively monitor for duplicated data. Trajectories that are identical lack the diversity required for robot learning and will not be credited by our scoring system. If we detect accounts submitting a massive volume of identical trajectories, all associated addresses will be permanently banned. For Axis, the quality and diversity of data are the only ways to solve the robotics generalization gap. They will always be our absolute top priorities.

Axis Robotics

30,876 görüntüleme • 4 ay önce

My focus is to make Mississippi the best state in the nation to live, work, and raise a family. Companies are investing here at a record pace. These investments mean new jobs. And these new jobs mean more opportunities for you to make money and better care for your family.

My focus is to make Mississippi the best state in the nation to live, work, and raise a family. Companies are investing here at a record pace. These investments mean new jobs. And these new jobs mean more opportunities for you to make money and better care for your family.

Governor Tate Reeves

37,954 görüntüleme • 3 yıl önce

Folks, here's another little design exercise that I spent the last few hours on 👍 This is based on a thesis I had for basically a decade now: Aiming in FPS games with the right analog stick is quite bad. Your thumb has maybe 1cm of travel on a tiny stick, fighting a spring, trying to make precise micro-adjustments AND fast 180-degree turns with the same input. If we're being honest, that's a terrible interface for the task. Your wrist, on the other hand, has roughly 180 degrees of rotational range, uses your entire forearm's muscle groups, and can make both ultra-precise micro-corrections and fast sweeping motions naturally. It's why mouse aiming destroys stick aiming: You're using your wrist and arm, not your thumb. So the thesis is simple: what if we just used wrist rotation directly as the aiming input? No VR headset, no special hardware. Just hold an old Switch1 Joy-Con in your right hand and rotate your wrist. Small rotations for precision tracking, larger rotations for fast turns. Recoil kicks the camera up and you instinctively tilt your hand back to compensate, etc. To me, it just feels right. I built a little prototype to test this. It has 4 weapons with some recoil, cover you can lean around, jump/crouch, lock-on targeting, ammo management, blablabla. The whole point is to see if this input method feels as good in practice as it does in theory. The aiming algorithm handles all the shitty stuff: Dead zones against shaky hands, auto-clutching when you unwind your wrist, asymmetric sensitivity because human wrists rotate further inward than outward, response curves for precision at low speed without sacrificing fast flicks, etc. I made a quick video to show the prototype in action. And if you wanna try it yourself, I put a small zip file up here that contains the executable:

Folks, here's another little design exercise that I spent the last few hours on 👍 This is based on a thesis I had for basically a decade now: Aiming in FPS games with the right analog stick is quite bad. Your thumb has maybe 1cm of travel on a tiny stick, fighting a spring, trying to make precise micro-adjustments AND fast 180-degree turns with the same input. If we're being honest, that's a terrible interface for the task. Your wrist, on the other hand, has roughly 180 degrees of rotational range, uses your entire forearm's muscle groups, and can make both ultra-precise micro-corrections and fast sweeping motions naturally. It's why mouse aiming destroys stick aiming: You're using your wrist and arm, not your thumb. So the thesis is simple: what if we just used wrist rotation directly as the aiming input? No VR headset, no special hardware. Just hold an old Switch1 Joy-Con in your right hand and rotate your wrist. Small rotations for precision tracking, larger rotations for fast turns. Recoil kicks the camera up and you instinctively tilt your hand back to compensate, etc. To me, it just feels right. I built a little prototype to test this. It has 4 weapons with some recoil, cover you can lean around, jump/crouch, lock-on targeting, ammo management, blablabla. The whole point is to see if this input method feels as good in practice as it does in theory. The aiming algorithm handles all the shitty stuff: Dead zones against shaky hands, auto-clutching when you unwind your wrist, asymmetric sensitivity because human wrists rotate further inward than outward, response curves for precision at low speed without sacrificing fast flicks, etc. I made a quick video to show the prototype in action. And if you wanna try it yourself, I put a small zip file up here that contains the executable:

thomasmahler

11,678 görüntüleme • 4 ay önce

I ran one of my own React dashboards through Replay before treating it as ready to ship. It looked fine to me. Replay QA found a notifications button that looked completely functional but did nothing when clicked, accessibility issues across the UI, and a workflow where completing a task failed to update the activity feed. Those are exactly the little things that are easy to miss when you're focused on building. With Replay for Teams, this kind of QA can also run around your team's GitHub workflow and surface debugging context as you ship. The app can look finished in the editor. What matters is what actually happens in the browser. Replay for Teams is launching today. Check it out on Product Hunt:

I ran one of my own React dashboards through Replay before treating it as ready to ship. It looked fine to me. Replay QA found a notifications button that looked completely functional but did nothing when clicked, accessibility issues across the UI, and a workflow where completing a task failed to update the activity feed. Those are exactly the little things that are easy to miss when you're focused on building. With Replay for Teams, this kind of QA can also run around your team's GitHub workflow and surface debugging context as you ship. The app can look finished in the editor. What matters is what actually happens in the browser. Replay for Teams is launching today. Check it out on Product Hunt:

Kaitee

57,558 görüntüleme • 5 gün önce

There is a major new update coming soon to Screen Studio! The update includes: Selfie camera layout: Choose the layout for your selfie camera during specific segments of your recording. You can display the camera in full screen or hide it completely. New cursor types: Expand your options beyond the default cursors. Playback speed settings: Enjoy support for the JKL playback controls. Faster exports: We've improved the export process for a better user experience. Additionally, there are several other small improvements and bug fixes. If you'd like to try the new features, you can download the latest beta version from We appreciate your feedback to help us thoroughly test the update. Stay tuned for even more updates to come!

There is a major new update coming soon to Screen Studio! The update includes: Selfie camera layout: Choose the layout for your selfie camera during specific segments of your recording. You can display the camera in full screen or hide it completely. New cursor types: Expand your options beyond the default cursors. Playback speed settings: Enjoy support for the JKL playback controls. Faster exports: We've improved the export process for a better user experience. Additionally, there are several other small improvements and bug fixes. If you'd like to try the new features, you can download the latest beta version from We appreciate your feedback to help us thoroughly test the update. Stay tuned for even more updates to come!

Screen Studio

62,035 görüntüleme • 2 yıl önce

Emergent is a platform that generates a complete app from a single sentence - including frontend and Python backend. No developers, no code, no long timelines. Whether it's a dashboard, SaaS tool or game: enter your idea and within minutes you have a working product. With the new update (V2.0), specialized agents for design, security review and testing have been integrated. This not only makes the creation process faster, but also significantly more robust. For founders, creators and teams that need speed, this is a real game changer! A great tool that comes at just the right time!

Emergent is a platform that generates a complete app from a single sentence - including frontend and Python backend. No developers, no code, no long timelines. Whether it's a dashboard, SaaS tool or game: enter your idea and within minutes you have a working product. With the new update (V2.0), specialized agents for design, security review and testing have been integrated. This not only makes the creation process faster, but also significantly more robust. For founders, creators and teams that need speed, this is a real game changer! A great tool that comes at just the right time!

Chubby♨️

32,357 görüntüleme • 1 yıl önce

I know everyone is prepping for the superbowl right now, but excited to share a new and improved AMM Visualizer, to help you debug and understand the strategies you create for the AMM challenge. Since I tweeted the original prototype, here's what's changed: - refactored from prototype into a full nextjs app - simulation now runs in a worker - added custom strategy support with in-browser compile + local save (no backend, no login). - added a full “per-pool Depth” view to compare your strategy vs Normalizer - added dark mode with theme toggle - added auto-zoom graph controls - improved curve/grid scaling behavior. - reworked into a full-screen terminal layout - made UX clearer overall - added Solidity syntax highlighting - improved responsive behavior on smaller screens

I know everyone is prepping for the superbowl right now, but excited to share a new and improved AMM Visualizer, to help you debug and understand the strategies you create for the AMM challenge. Since I tweeted the original prototype, here's what's changed: - refactored from prototype into a full nextjs app - simulation now runs in a worker - added custom strategy support with in-browser compile + local save (no backend, no login). - added a full “per-pool Depth” view to compare your strategy vs Normalizer - added dark mode with theme toggle - added auto-zoom graph controls - improved curve/grid scaling behavior. - reworked into a full-screen terminal layout - made UX clearer overall - added Solidity syntax highlighting - improved responsive behavior on smaller screens

Andrea | Devrelius

16,553 görüntüleme • 6 ay önce

The world's smallest industrial robot arm Built for sub-micrometer precision. Here’s a behind-the-scenes look at how to control what is likely the world’s smallest and most precise industrial robot arm using custom software. The stack includes online and offline programming with simulation, an analytical kinematics solver, a singularity-aware trajectory planner, and an API layer for integration with external systems. It’s designed for applications where positioning error is measured in micrometers, not millimeters. Microelectronics, semiconductor handling, photonics alignment, medical device assembly, lab automation. When parts are tiny and tolerances are unforgiving, the control architecture matters as much as the mechanics. This is from Oleksander Stepanenko (Oleksandr Stepanenko), if you don’t follow him and his YouTube channel… no one can help you. ——- Weekly robotics and AI insights. Subscribe free:

The world's smallest industrial robot arm Built for sub-micrometer precision. Here’s a behind-the-scenes look at how to control what is likely the world’s smallest and most precise industrial robot arm using custom software. The stack includes online and offline programming with simulation, an analytical kinematics solver, a singularity-aware trajectory planner, and an API layer for integration with external systems. It’s designed for applications where positioning error is measured in micrometers, not millimeters. Microelectronics, semiconductor handling, photonics alignment, medical device assembly, lab automation. When parts are tiny and tolerances are unforgiving, the control architecture matters as much as the mechanics. This is from Oleksander Stepanenko (Oleksandr Stepanenko), if you don’t follow him and his YouTube channel… no one can help you. ——- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

60,285 görüntüleme • 5 ay önce

Tested ChatGPT's new Shopping Research feature and I'm pretty impressed. The UI is adaptive to what you're searching for - it asked me different questions and even organized the results differently when I was looking for "Christmas gift for a 13 year old girl" vs. "best white sneakers for work". It also asks you to rate products while conducting the search to better refine options (with links and reviews provided) - this makes the wait time more tolerable. And IMO, the results are quite good, with detailed justifications for each product. This is a big step up from the other products we've seen in this space (ex. Perplexity Shopping) - I'm curious if they double down on building out UI here...

Tested ChatGPT's new Shopping Research feature and I'm pretty impressed. The UI is adaptive to what you're searching for - it asked me different questions and even organized the results differently when I was looking for "Christmas gift for a 13 year old girl" vs. "best white sneakers for work". It also asks you to rate products while conducting the search to better refine options (with links and reviews provided) - this makes the wait time more tolerable. And IMO, the results are quite good, with detailed justifications for each product. This is a big step up from the other products we've seen in this space (ex. Perplexity Shopping) - I'm curious if they double down on building out UI here...

Olivia Moore

53,447 görüntüleme • 9 ay önce

Pony Waifu Sim - Alpha 5.5 Release! Happy New Year everyone! We are starting off 2024 with the release of the Memory Mirror update! This update brings with it a feature that has been highly requested: The ability to replay lewd scenes! Simply spend the night in your bed in order to receive a visit from a strange, yet familiar, presence. There might even be something special for those who are particularly lonely. The new scene features our first musical track from the amazing BGM as well! The builds are available two days early on our Patreon and SubscribeStar pages! Patch Notes: - Added the Memory Mirror and its accompanying story. - A very special surprise for the loneliest of players. (You will likely need to start a new save for this.) - Fixed an issue preventing players from reading books, which is required for progressing Purple’s route. - Many other minor bugfixes.

Sensitive content

Pony Waifu Sim - Alpha 5.5 Release! Happy New Year everyone! We are starting off 2024 with the release of the Memory Mirror update! This update brings with it a feature that has been highly requested: The ability to replay lewd scenes! Simply spend the night in your bed in order to receive a visit from a strange, yet familiar, presence. There might even be something special for those who are particularly lonely. The new scene features our first musical track from the amazing BGM as well! The builds are available two days early on our Patreon and SubscribeStar pages! Patch Notes: - Added the Memory Mirror and its accompanying story. - A very special surprise for the loneliest of players. (You will likely need to start a new save for this.) - Fixed an issue preventing players from reading books, which is required for progressing Purple’s route. - Many other minor bugfixes.

StudioWhy

42,217 görüntüleme • 2 yıl önce

Have you ever felt a big difference in "aiming feel" between two mice, that you couldn't fully justify w/ just the shape or DPI deviation? THIS might be the reason. You might not be accounting for how the sensor position relative to your wrist differs between the two. But on FMM, you can now convert your sensitivity across games AND mice! I've explained before that when I got my original world records in 2020/2021 for static aiming, I thought that a lower sensor position on some mice was a big reason for it, as I could have lower-sensitivity micro corrections while maintaining a higher cm/360 for the large arm swipes. I've also explained it before when analyzing Donk's grip here: The difference can be MASSIVE. Assuming you grip the GPX2 and ATK F1 Extreme the same way, here's an example (taken straight from the converter) of how large the difference is (check the video!). 3.7@800 dpi in OW2 w/ the F1 feels the same as 3.33@800 dpi on the GPX2 X2 when using your wrist. If you want to maintain a CONSISTENT wrist feel, then you need to adjust the sensitivity around this when changing mice. The cm/360 WILL differ, so your arm sensitivity, which we can consider to be purely "translational", will be different, but I think that most of what we "feel" is actually how the mouse responds to wrist movements when flicking or micro correcting. Let me know what y'all think and if you find this useful! The feature might be a bit complex at first, but hopefully it's understandable. BONUS POINT, I also added a little table that highlights if your given sens in-game at your DPI will incur in pixel skipping, so you can decide to increase DPI at your resolution/sensitivity, which I've also talked about many times before. Stay tuned as I have a ton more cool stuff coming for FMM!

Have you ever felt a big difference in "aiming feel" between two mice, that you couldn't fully justify w/ just the shape or DPI deviation? THIS might be the reason. You might not be accounting for how the sensor position relative to your wrist differs between the two. But on FMM, you can now convert your sensitivity across games AND mice! I've explained before that when I got my original world records in 2020/2021 for static aiming, I thought that a lower sensor position on some mice was a big reason for it, as I could have lower-sensitivity micro corrections while maintaining a higher cm/360 for the large arm swipes. I've also explained it before when analyzing Donk's grip here: The difference can be MASSIVE. Assuming you grip the GPX2 and ATK F1 Extreme the same way, here's an example (taken straight from the converter) of how large the difference is (check the video!). 3.7@800 dpi in OW2 w/ the F1 feels the same as 3.33@800 dpi on the GPX2 X2 when using your wrist. If you want to maintain a CONSISTENT wrist feel, then you need to adjust the sensitivity around this when changing mice. The cm/360 WILL differ, so your arm sensitivity, which we can consider to be purely "translational", will be different, but I think that most of what we "feel" is actually how the mouse responds to wrist movements when flicking or micro correcting. Let me know what y'all think and if you find this useful! The feature might be a bit complex at first, but hopefully it's understandable. BONUS POINT, I also added a little table that highlights if your given sens in-game at your DPI will incur in pixel skipping, so you can decide to increase DPI at your resolution/sensitivity, which I've also talked about many times before. Stay tuned as I have a ton more cool stuff coming for FMM!

bardOZ (Giovanni Laerte Frongia)

402,417 görüntüleme • 3 ay önce

Episode 242: Khala, Collective Intelligence We release our new flagship product in research preview. Khala is the first open-source orchestrator of models, agents, tools, skills & programs behind a familiar OpenAI-compatible API endpoint. Unlike the closed orchestrators (OpenRouter Fusion, Sakana Fugu): ours is fully open end to end; every call is metered and inspectable — you can see what ran, what it cost, and who did the work; and it's built to pay them, in bitcoin, for the compute, data, labor, and verification they contribute. Under the hood it's our Blueprint system — DSPy for the orchestrator: every module is its own signature, independently optimizable via GEPA and independently discoverable + monetizable on our marketplace. Improve one piece, the whole collective gets better. And it improves instead of depreciating: our Tassadar training run is how the collective trains its own open models and compiles verifiable-by-replay specialists that plug straight into Khala. We are thrilled to begin organizing the world's intelligence into one open network accessible to all people and agents. Try it for free 👉

Episode 242: Khala, Collective Intelligence We release our new flagship product in research preview. Khala is the first open-source orchestrator of models, agents, tools, skills & programs behind a familiar OpenAI-compatible API endpoint. Unlike the closed orchestrators (OpenRouter Fusion, Sakana Fugu): ours is fully open end to end; every call is metered and inspectable — you can see what ran, what it cost, and who did the work; and it's built to pay them, in bitcoin, for the compute, data, labor, and verification they contribute. Under the hood it's our Blueprint system — DSPy for the orchestrator: every module is its own signature, independently optimizable via GEPA and independently discoverable + monetizable on our marketplace. Improve one piece, the whole collective gets better. And it improves instead of depreciating: our Tassadar training run is how the collective trains its own open models and compiles verifiable-by-replay specialists that plug straight into Khala. We are thrilled to begin organizing the world's intelligence into one open network accessible to all people and agents. Try it for free 👉

OpenAgents

238,777 görüntüleme • 1 ay önce

We just launched our new website along with some great changes: 1. Added /themes page where you can grab our free shadcn / ui themes. 2. The /academy is now free for everyone. You can learn the Figma to shadcn/ui process with our short video guides. 3. The site is rebuilt from the ground up with Next.js 15, React 19, and shadcn/ui. The new stack gives us faster load times, better accessibility, and a more cohesive experience across all pages. I'm already working on more themes, expanded docs, and new Pro Blocks. Thanks for being here and supporting our small project! More to come!

We just launched our new website along with some great changes: 1. Added /themes page where you can grab our free shadcn / ui themes. 2. The /academy is now free for everyone. You can learn the Figma to shadcn/ui process with our short video guides. 3. The site is rebuilt from the ground up with Next.js 15, React 19, and shadcn/ui. The new stack gives us faster load times, better accessibility, and a more cohesive experience across all pages. I'm already working on more themes, expanded docs, and new Pro Blocks. Thanks for being here and supporting our small project! More to come!

Matt

44,889 görüntüleme • 10 ay önce

Introducing #DJIRS5, a lightweight commercial stabilizer built to lead the scene. Featuring touchscreen subject selection with the new Enhanced Intelligent Tracking Module, enabling precise tracking of people, vehicles, pets, and custom subjects. All-axis fine-tuning knobs deliver smooth, accurate balancing, while the Electronic Briefcase Handle offers direct, intuitive control for more creative angles. Faster charging and extended battery life keep creativity moving, turning ambition into cinematic results. Key features of the DJI RS 5: ✅ Enhanced Intelligent Tracking and Composition ✅ Briefcase Handle With Gimbal & Camera Control ✅ 1-Hour Full Charge, 15% Extended Runtime ✅ Fine-Tuning Knobs and Teflon™ Enhanced Balancing ✅ 2nd-Gen Automated Axis Locks ✅ New Z-Axis Indicator Learn more: Buy now: *The Enhanced Intelligent Tracking Module and Electronic Briefcase Handle are available for separate purchase or as part of a compatible combo. *The DJI RS BG33 Battery Grip is not available for purchase in India. #DJI #DJIRonin #DJIRS5 #RS5 #CameraGimbal #CameraStabilizer #Videography

Introducing #DJIRS5, a lightweight commercial stabilizer built to lead the scene. Featuring touchscreen subject selection with the new Enhanced Intelligent Tracking Module, enabling precise tracking of people, vehicles, pets, and custom subjects. All-axis fine-tuning knobs deliver smooth, accurate balancing, while the Electronic Briefcase Handle offers direct, intuitive control for more creative angles. Faster charging and extended battery life keep creativity moving, turning ambition into cinematic results. Key features of the DJI RS 5: ✅ Enhanced Intelligent Tracking and Composition ✅ Briefcase Handle With Gimbal & Camera Control ✅ 1-Hour Full Charge, 15% Extended Runtime ✅ Fine-Tuning Knobs and Teflon™ Enhanced Balancing ✅ 2nd-Gen Automated Axis Locks ✅ New Z-Axis Indicator Learn more: Buy now: The Enhanced Intelligent Tracking Module and Electronic Briefcase Handle are available for separate purchase or as part of a compatible combo. The DJI RS BG33 Battery Grip is not available for purchase in India. #DJI #DJIRonin #DJIRS5 #RS5 #CameraGimbal #CameraStabilizer #Videography

DJI

62,155 görüntüleme • 6 ay önce

You can now try Llama 3.1 405B for free (link below)! This is the largest open-source model out there, and for the first time, an open model is competitive with closed models. This time around, Meta did something new: Llama 3.1 has a license that allows developers to use it to enhance other models. For the first time, you can distill Llama 3.1 405B's capabilities into a smaller, more practical model for your use case. First, here is the link where you can play with Llama 3.1 for free: The model is hosted in Tune Studio, an end-to-end platform for developing applications using Large Language Models. They are sponsoring this post. Take a look at the attached video. It will show you how you can fine-tune a simple model using Llama 3.1 without leaving the platform: 1. You can create an empty dataset 2. Use the playground to generate and record interactions with Llama 3.1 3. Modify the dataset directly using the playground 4. Export the data and fine-tune a smaller model Fast and easy! As long as you have a web browser, you can start experimenting with fine-tuning and Llama 3.1. That's all it takes!

You can now try Llama 3.1 405B for free (link below)! This is the largest open-source model out there, and for the first time, an open model is competitive with closed models. This time around, Meta did something new: Llama 3.1 has a license that allows developers to use it to enhance other models. For the first time, you can distill Llama 3.1 405B's capabilities into a smaller, more practical model for your use case. First, here is the link where you can play with Llama 3.1 for free: The model is hosted in Tune Studio, an end-to-end platform for developing applications using Large Language Models. They are sponsoring this post. Take a look at the attached video. It will show you how you can fine-tune a simple model using Llama 3.1 without leaving the platform: 1. You can create an empty dataset 2. Use the playground to generate and record interactions with Llama 3.1 3. Modify the dataset directly using the playground 4. Export the data and fine-tune a smaller model Fast and easy! As long as you have a web browser, you can start experimenting with fine-tuning and Llama 3.1. That's all it takes!

Santiago

55,609 görüntüleme • 2 yıl önce