正在加载视频...

视频加载失败

返回首页

Train AI robots without writing a single line of code. 🤖 We just launched LeLab, the official graphical user interface for LeRobot built by Nicolas Rabault. It completely removes the command line from the robot learning workflow, taking you from raw hardware to autonomous movement visually. If you've ever...

LeRobotHF's profile picture

LeRobot

17,439 subscribers

49,744 次观看 • 28 天前 •via X (Twitter)

教育科学技术体育
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:
0:29

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

42,804 次观看 • 5 个月前

Tired of teleoperating your robots? We built a way to scale robot datasets without teleop, dynamic simulation, or even robot hardware. Just one smartphone scan + one human hand demo video → thousands of diverse robot trajectories. Trainable by diffusion policy and VLA models as-is. Introducing: Real2Render2Real 👉
0:48

Tired of teleoperating your robots? We built a way to scale robot datasets without teleop, dynamic simulation, or even robot hardware. Just one smartphone scan + one human hand demo video → thousands of diverse robot trajectories. Trainable by diffusion policy and VLA models as-is. Introducing: Real2Render2Real 👉

Max Fu

69,310 次观看 • 1 年前

What if your robot could walk into a café, place an order for you, and pay — all by itself? With x402, that future is now. No accounts. No API keys. Just one line of code for AI agents to make seamless, autonomous payments. Built with Coinbase Developer Platform🛡️'s AgentKit, brought to life by OpenMind 🤖☕
2:46

What if your robot could walk into a café, place an order for you, and pay — all by itself? With x402, that future is now. No accounts. No API keys. Just one line of code for AI agents to make seamless, autonomous payments. Built with Coinbase Developer Platform🛡️'s AgentKit, brought to life by OpenMind 🤖☕

OpenMind

124,143 次观看 • 1 年前

Open-source robot arm meets hand tracking [📍GitHub below] It is designed with an industrial mindset but built as a 3D-printed desktop system. PAROL6 paired with a LEAP Motion controller is a nice example of how accessible robot teleoperation has become. • Hand motion is streamed to the robot at 100 Hz via UDP • A pneumatic gripper is controlled by simple fist open and close gestures • The entire robot stack is open source, from mechanics to control software Combine that with low-latency hand tracking and you get a very practical platform for learning manipulation, teleoperation, and human-robot interfaces. This kind of setup is great for experimentation, teleop, data collection, and teaching robots by demonstration All without proprietary hardware or locked software. Credit to SourceRobotics 📍Code: —— Weekly robotics and AI insights. Subscribe free:
0:35

Open-source robot arm meets hand tracking [📍GitHub below] It is designed with an industrial mindset but built as a 3D-printed desktop system. PAROL6 paired with a LEAP Motion controller is a nice example of how accessible robot teleoperation has become. • Hand motion is streamed to the robot at 100 Hz via UDP • A pneumatic gripper is controlled by simple fist open and close gestures • The entire robot stack is open source, from mechanics to control software Combine that with low-latency hand tracking and you get a very practical platform for learning manipulation, teleoperation, and human-robot interfaces. This kind of setup is great for experimentation, teleop, data collection, and teaching robots by demonstration All without proprietary hardware or locked software. Credit to SourceRobotics 📍Code: —— Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

32,025 次观看 • 5 个月前

We asked Angus about the future of robotics and AI training. “I built an industrial-grade kinematic solver for 6 degrees of freedom motion, with full path planning and joint control." "You can stream AI model outputs into it, or run it like a traditional industrial robot." "The benefit of that is all of that information you can stream and record from those joints… you can use that as the training data." "Frankly, moving a robot with a lever is really crap data for training robots. But this gives you smooth, high-quality motion, without needing two robots.”
2:16

We asked Angus about the future of robotics and AI training. “I built an industrial-grade kinematic solver for 6 degrees of freedom motion, with full path planning and joint control." "You can stream AI model outputs into it, or run it like a traditional industrial robot." "The benefit of that is all of that information you can stream and record from those joints… you can use that as the training data." "Frankly, moving a robot with a lever is really crap data for training robots. But this gives you smooth, high-quality motion, without needing two robots.”

TBPN

18,811 次观看 • 1 年前

Not a dev. Not a researcher. We’re just a human with internet access. So how do we even help AI get better? PublicAI lets you help train AI, and actually earn from it! 🎙️ Talk. Type. Share content. 🧠 Your data = training fuel 💸 You get a cut. This is how the next wave of AI is being built: by all of us.
0:51

Not a dev. Not a researcher. We’re just a human with internet access. So how do we even help AI get better? PublicAI lets you help train AI, and actually earn from it! 🎙️ Talk. Type. Share content. 🧠 Your data = training fuel 💸 You get a cut. This is how the next wave of AI is being built: by all of us.

Bibs 🎮

165,680 次观看 • 1 年前

I have one of the largest collection of process films known for AI training and robot training. Most have only a single copy remaining. This is how you get to first principles. 1975, the Short Order Cook…
2:58

I have one of the largest collection of process films known for AI training and robot training. Most have only a single copy remaining. This is how you get to first principles. 1975, the Short Order Cook…

Brian Roemmele

15,218 次观看 • 1 个月前

🚀 We just shipped a big upgrade to our imitation-learning-in-simulation playground in LeRobot, built together with Lightwheel ! You can now teleoperate robots in sim (keyboard or real robot) and collect training demos instantly. This makes it possible to run real IL research on harder, more realistic manipulation tasks, even if you don’t have hardware. New tasks: 🟠 pick orange to the plate 🧺 fold cloth (yes!) 📦 pick 2 “e” toys to the box 🔴 lift red cube Through our partnership with Lightwheel, LeIsaac was integrated into EnvHub on day one, a best-practice integration that strengthens both ecosystems and pushes simulation-first robotics forward. 👉 Load a task via EnvHub, start teleop-ing, record your demo, upload the data on the hub and you’re ready to train or test.
0:22

🚀 We just shipped a big upgrade to our imitation-learning-in-simulation playground in LeRobot, built together with Lightwheel ! You can now teleoperate robots in sim (keyboard or real robot) and collect training demos instantly. This makes it possible to run real IL research on harder, more realistic manipulation tasks, even if you don’t have hardware. New tasks: 🟠 pick orange to the plate 🧺 fold cloth (yes!) 📦 pick 2 “e” toys to the box 🔴 lift red cube Through our partnership with Lightwheel, LeIsaac was integrated into EnvHub on day one, a best-practice integration that strengthens both ecosystems and pushes simulation-first robotics forward. 👉 Load a task via EnvHub, start teleop-ing, record your demo, upload the data on the hub and you’re ready to train or test.

LeRobot

44,647 次观看 • 7 个月前

#mixtral #mistral #LLM360 Serving Mixtral and LLM360 on FEDML Nexus AI ( We offer Mixtral model endpoints the cheapest in the market: only $0.0005 / 1K tokens! FEDML embraces open source and open model weights. We believe the future of AI belongs to large-scale open collaboration. Today we are excited to support new advances in open-source foundation models: Mixtral, the latest open-source LLM beating Llama2-70B with Mixture-of-Experts (MoE) architecture, and Amber and CrystalCoder backed by LLM360, the framework for open-source LLMs to foster transparency, trust, and collaborative research. Compared to existing fragmented ML products in the market, FEDML Nexus AI is the next-gen cloud service for LLM and Generative AI. It provides an end-to-end platform backed by serverless/decentralized AI infrastructure. Specifically: 1. Economical Serving Engine, ScaleLLM, is where you run your model in cheaper price by optimizing GPU memory and with fully optimized throughput for supporting more concurrent requests. 2. FEDML® Deploy simplifies CLI and MLOps workflow for model deployment on a serverless GPU cloud or on-premise cluster. 3. Serverless Endpoint runs on serverless GPU clouds. With our pay per use policy, we abstract the responsibility of acquiring or leasing an extensive GPU inventory when your are uncertain about your future AI service traffic. The autoscaling feature seamlessly adjusts the backend GPU resources in response to your service traffic. 4. On-premise Deployment helps you own your LLM model on your local environment with AI safety support. 5. FEDML® Launch for serverless GPU clouds. With one-line CLI, it swiftly pairs AI jobs with the most economical GPU resources, auto-provisions, and effortlessly runs the job, abstracting complex environment setup and management. 6. Zero-code Fine-tuning supported by FEDML® Studio optimizes your model on your domain-specific data without writing any line of source code. 7. Pre-training LLM supports cluster management and experimental tracking. You maintain your training clusters for your urgent needs in your vertical domain. As a closing note, FEDML is gearing up to unveil a cutting-edge service for LLM-based agents and our own cost-effective LLM. Please stay tuned and keep an eye out for upcoming announcements!
0:30

#mixtral #mistral #LLM360 Serving Mixtral and LLM360 on FEDML Nexus AI ( We offer Mixtral model endpoints the cheapest in the market: only $0.0005 / 1K tokens! FEDML embraces open source and open model weights. We believe the future of AI belongs to large-scale open collaboration. Today we are excited to support new advances in open-source foundation models: Mixtral, the latest open-source LLM beating Llama2-70B with Mixture-of-Experts (MoE) architecture, and Amber and CrystalCoder backed by LLM360, the framework for open-source LLMs to foster transparency, trust, and collaborative research. Compared to existing fragmented ML products in the market, FEDML Nexus AI is the next-gen cloud service for LLM and Generative AI. It provides an end-to-end platform backed by serverless/decentralized AI infrastructure. Specifically: 1. Economical Serving Engine, ScaleLLM, is where you run your model in cheaper price by optimizing GPU memory and with fully optimized throughput for supporting more concurrent requests. 2. FEDML® Deploy simplifies CLI and MLOps workflow for model deployment on a serverless GPU cloud or on-premise cluster. 3. Serverless Endpoint runs on serverless GPU clouds. With our pay per use policy, we abstract the responsibility of acquiring or leasing an extensive GPU inventory when your are uncertain about your future AI service traffic. The autoscaling feature seamlessly adjusts the backend GPU resources in response to your service traffic. 4. On-premise Deployment helps you own your LLM model on your local environment with AI safety support. 5. FEDML® Launch for serverless GPU clouds. With one-line CLI, it swiftly pairs AI jobs with the most economical GPU resources, auto-provisions, and effortlessly runs the job, abstracting complex environment setup and management. 6. Zero-code Fine-tuning supported by FEDML® Studio optimizes your model on your domain-specific data without writing any line of source code. 7. Pre-training LLM supports cluster management and experimental tracking. You maintain your training clusters for your urgent needs in your vertical domain. As a closing note, FEDML is gearing up to unveil a cutting-edge service for LLM-based agents and our own cost-effective LLM. Please stay tuned and keep an eye out for upcoming announcements!

TensorOpera AI

90,271 次观看 • 2 年前

Our folding project showed what is possible with open-source. Reproducing it shouldn't take a PhD. LeRobot now ships with AGENT_GUIDE.md → so now your AI agent knows the insights, how to set up an SO-101, record good data, and pick a policy for your GPU. Just ask!
0:36

Our folding project showed what is possible with open-source. Reproducing it shouldn't take a PhD. LeRobot now ships with AGENT_GUIDE.md → so now your AI agent knows the insights, how to set up an SO-101, record good data, and pick a policy for your GPU. Just ask!

LeRobot

30,638 次观看 • 2 个月前

Your AI, Now Powered by Your Own Computer. 🚀 What if your AI could actually do the work on your laptop? Claw Intelligence makes it possible. 1️⃣ Link: Register your device with a simple terminal command. 2️⃣ Command: Tell the AI to handle file transfers or environment setups. 3️⃣ Execute: Watch as the AI prepares your local workspace for you. Check out the video to see how we sync with your computer to streamline a BTC monitor task! 📈 Tyr Now:
0:56

Your AI, Now Powered by Your Own Computer. 🚀 What if your AI could actually do the work on your laptop? Claw Intelligence makes it possible. 1️⃣ Link: Register your device with a simple terminal command. 2️⃣ Command: Tell the AI to handle file transfers or environment setups. 3️⃣ Execute: Watch as the AI prepares your local workspace for you. Check out the video to see how we sync with your computer to streamline a BTC monitor task! 📈 Tyr Now:

Claw Intelligence

12,323 次观看 • 2 个月前

AI Decides if You Can Drive Asking a frickin robot if you can drive your car is about to be our new reality here in Garbageville. 57 Clownpublicans just gave some AI terminator the ability to shut off your car. We started by telling a website robot that we are not a robot and now we're about to be negotiating the ability to drive with with ChatGARB. Glad the gubmint is focusing on things we need and stuff. No big deal! Consume your slop, Thomas Massie is a gRanDStAnDA and 57 goons just clapped your ability to drive, by decree, for a technology that doesn't even exist yet. Nothing to see here! What could go wrong? Sacrifice liberty for fake safety, the murican way!
2:49

AI Decides if You Can Drive Asking a frickin robot if you can drive your car is about to be our new reality here in Garbageville. 57 Clownpublicans just gave some AI terminator the ability to shut off your car. We started by telling a website robot that we are not a robot and now we're about to be negotiating the ability to drive with with ChatGARB. Glad the gubmint is focusing on things we need and stuff. No big deal! Consume your slop, Thomas Massie is a gRanDStAnDA and 57 goons just clapped your ability to drive, by decree, for a technology that doesn't even exist yet. Nothing to see here! What could go wrong? Sacrifice liberty for fake safety, the murican way!

Bobby Sauce

18,063 次观看 • 2 个月前

With just a one line installation, qualify for Billions FAIAR reward. As Billions has introduced the First AI Agent Rewards and It’s available for only verified agents. No complex setup, Just Termux(Android)/Mac/Windows + one command ✅ Get the one line installation with the link below 👇 Link: NOTE: This also account if you haven't gotten an agent linked to your Billions account or if you don't have the private key to your Billions agent and your machine is wiped. Check the video to understand better. Don't snooze, get your agent earning now with billions
4:55

With just a one line installation, qualify for Billions FAIAR reward. As Billions has introduced the First AI Agent Rewards and It’s available for only verified agents. No complex setup, Just Termux(Android)/Mac/Windows + one command ✅ Get the one line installation with the link below 👇 Link: NOTE: This also account if you haven't gotten an agent linked to your Billions account or if you don't have the private key to your Billions agent and your machine is wiped. Check the video to understand better. Don't snooze, get your agent earning now with billions

fashaking 🦇

24,939 次观看 • 2 个月前

Excited to announce Tau Robotics (Tau Robotics). We are building a general AI for robots. We start by building millions of robot arms that learn in the real world. In the video, two robot arms are fully autonomous and controlled by a single neural network conditioned on different language instructions (four axes and five axes robot arms). The other two arms are teleoperated. The entire hardware cost in the video is about $1400. The video is at 1.5x speed.
0:39

Excited to announce Tau Robotics (Tau Robotics). We are building a general AI for robots. We start by building millions of robot arms that learn in the real world. In the video, two robot arms are fully autonomous and controlled by a single neural network conditioned on different language instructions (four axes and five axes robot arms). The other two arms are teleoperated. The entire hardware cost in the video is about $1400. The video is at 1.5x speed.

Alexander Koch

437,797 次观看 • 2 年前

How to Convert Your ComfyUI Workflow into an APP Mode Workflow App Mode is a new way to view and run any ComfyUI workflow. When you enter App Mode, the node graph disappears and is replaced by a clean, purpose-built interface: just the inputs your user needs.
4:04

How to Convert Your ComfyUI Workflow into an APP Mode Workflow App Mode is a new way to view and run any ComfyUI workflow. When you enter App Mode, the node graph disappears and is replaced by a clean, purpose-built interface: just the inputs your user needs.

ComfyUI

79,144 次观看 • 1 个月前

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.
1:00

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.

Neuracore

40,620 次观看 • 7 个月前

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵
0:58

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵

Jim Fan

165,246 次观看 • 1 年前

We’re thrilled to announce the Antigravity CLI, a lightweight way to spin up the same Antigravity agents right from the terminal. 💻 It gives you the exact same harness and same models, with a product experience tailored for the command line. It adapts entirely to you: your keybindings, your themes, your workflows. Learn how to get started with the Antigravity CLI: 👇
3:03

We’re thrilled to announce the Antigravity CLI, a lightweight way to spin up the same Antigravity agents right from the terminal. 💻 It gives you the exact same harness and same models, with a product experience tailored for the command line. It adapts entirely to you: your keybindings, your themes, your workflows. Learn how to get started with the Antigravity CLI: 👇

Google Antigravity

116,227 次观看 • 1 个月前

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork
0:20

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork

Pascal Bornet

185,867 次观看 • 2 个月前