Everyone Can Build a Robot: Open Source Embodied AI With Seeed Studio | NVIDIA AI Podcast Ep. 300
Episode
29 min
Read time
2 min
Topics
Startups, Fundraising & VC, Leadership
AI-Generated Summary
Key Takeaways
- ✓Affordable Entry Point: Seeed Studio's SO-ARM robot arm costs $200 and ships ready to train without programming knowledge. Users teach it tasks by physically guiding its movements several times, sending that motion data to cloud training, then deploying the resulting model back onto the device — reducing onboarding from months to days.
- ✓Imitation-Based Robot Training: Rather than coding spatial planning algorithms, users now train robot arms the way they train animals — physically demonstrating a task repeatedly, uploading the recorded data for cloud-based model training, then deploying via Jetson. This shifts robot programming from engineering expertise to domain expertise, letting chefs or craftspeople train their own robots.
- ✓OpenClaw + Jetson Local Deployment: Installing OpenClaw locally on a Jetson device enables natural language robot control — typing "move arm up" or "pick up object" executes physical commands without writing code. Running the Qwen 3.5 model locally means no API tokens or cloud costs, making autonomous robot operation viable for small businesses under $1,000.
- ✓Modular Robot Architecture Over Humanoids: Rather than building general-purpose humanoid robots, Seeed Studio produces interchangeable components — arms, chassis, hands, torsos, and wheeled bases — that developers combine for specific use cases. This approach lets startups and researchers build domain-specific robots faster, with Seeed providing manufacturing scale-up from prototype to 3,000 shipped units in five months.
- ✓Sim-to-Real Gap Closing via NVIDIA Isaac Sim: Developers can mirror Seeed robot models inside NVIDIA Isaac Sim to validate and tune robot behavior before physical deployment. As affordable hardware expands the user base, more people contribute real-world training data, accelerating the feedback loop between simulation accuracy and physical performance across diverse environments.
What It Covers
Seeed Studio CEO Eric Pan and robotics head Elaine Wu explain how their open-source hardware company, operating since 2008, is democratizing physical AI through affordable robot arms starting at $200, Jetson-powered edge computing, and OpenClaw integration that lets users control robots via text commands.
Key Questions Answered
- •Affordable Entry Point: Seeed Studio's SO-ARM robot arm costs $200 and ships ready to train without programming knowledge. Users teach it tasks by physically guiding its movements several times, sending that motion data to cloud training, then deploying the resulting model back onto the device — reducing onboarding from months to days.
- •Imitation-Based Robot Training: Rather than coding spatial planning algorithms, users now train robot arms the way they train animals — physically demonstrating a task repeatedly, uploading the recorded data for cloud-based model training, then deploying via Jetson. This shifts robot programming from engineering expertise to domain expertise, letting chefs or craftspeople train their own robots.
- •OpenClaw + Jetson Local Deployment: Installing OpenClaw locally on a Jetson device enables natural language robot control — typing "move arm up" or "pick up object" executes physical commands without writing code. Running the Qwen 3.5 model locally means no API tokens or cloud costs, making autonomous robot operation viable for small businesses under $1,000.
- •Modular Robot Architecture Over Humanoids: Rather than building general-purpose humanoid robots, Seeed Studio produces interchangeable components — arms, chassis, hands, torsos, and wheeled bases — that developers combine for specific use cases. This approach lets startups and researchers build domain-specific robots faster, with Seeed providing manufacturing scale-up from prototype to 3,000 shipped units in five months.
- •Sim-to-Real Gap Closing via NVIDIA Isaac Sim: Developers can mirror Seeed robot models inside NVIDIA Isaac Sim to validate and tune robot behavior before physical deployment. As affordable hardware expands the user base, more people contribute real-world training data, accelerating the feedback loop between simulation accuracy and physical performance across diverse environments.
Notable Moment
When Seeed connected their robot arm to OpenClaw two weeks before recording, they gave it no explicit instructions beyond finding its own libraries and placing itself into the physical world — and it independently planned how to interpret movement commands like "move ten centimeters up."
Episode Transcript
Previously, you need to spend months of trainings Yeah. To understand the spiritual, like, planning on the robots, how it moves. But now what you do with the robots is after they're sitting up, you train it. Yeah. Like you train a dog. You teach it how to do it with by holding its hand to do the operations for several times, then you send all the data back to train on the cloud and deploy on adjacent. Welcome to the NVIDIA AI podcast. I'm Noah Kravitz. Our guests today are Elaine Wu and Eric Pan from Seed Studio. Eric is the CEO. Elaine is the head of robotics. And if you're not watching this on video, if you're listening on audio, you might wanna check out the video stream. We've got a handful of robot companions with us today. We're gonna get some demos. We're gonna talk about Open Robotics, OpenClaw, edge computing, robotics, and AI, all kinds of great stuff. So let's get into it. Eric, Elaine, welcome. Thanks for joining the NVIDIA AI podcast. Pleasure. Thank you for inviting us. Thank you for having us. And thank you for bringing your friends. So maybe we can start, and Eric, I'll ask you first, for a little bit of an overview for listeners, viewers who might not know Seed Studio, what the company is and how you moved from open hardware into robotics and edge AI. Sure. So at Seed Studio, we have been supporting the global communities on open source hardware from 2008 for eighteen years. So we do starting by open source hardware modules, but we move on to all kinds of devices. We work with NVIDIA for several years, around just nanos, but now we have a lot of more robotics, open source robotics from arms, from little desktop, embodiments. So this goes on. Yeah. So as, NVIDIA, elite partner, so I see that our AI robotics team have built a lineup of the Jetson O ring devices as the most powerful and advanced robot brain. Together with that, we also have a full stack of robotic components from perception and from the control. So I see that our mission is make technology more accessible to everyone. And so how is this convergence of AI and OpenAI and robotics hardware and technologies now like OpenClaw? How is this shaping the world of robotics and and open robotics? I think it's a very, evolving and progressive, trend. If you're starting to understand microcontrollers, okay, everybody can use microcontrollers through all the open source hardware and make a movement. Then gradually people realize, okay, I have open source hardware. Can I use it for industries? Can I do real projects with that? So you see a lot of IoT, a lot of, robotics in early times start to pop up because it's easier for everyone cross boundaries coming from different industries. Now they can use technology for themselves no matter where they are all over the world. …
Get the full transcript (4,968 words) + summary by email — free
One-time email with the complete transcript and AI summary of this episode. No account needed.
One email, no spam. We’ll also show you what SignalCast does.
You just read a 3-minute summary of a 26-minute episode.
Get NVIDIA AI Podcast summarized like this every Monday — plus up to 2 more podcasts, free.
Pick Your Podcasts — FreeKeep Reading
More from NVIDIA AI Podcast
Inside Instacart's AI-Powered Smart Shopping Cart | NVIDIA AI Podcast Ep. 302
Jun 24 · 39 min
Accidental Tech Podcast
686: Write Two Letters
Apr 9
More from NVIDIA AI Podcast
How Mistral Is Building Frontier AI for the Enterprise | NVIDIA AI Podcast Ep. 301
Jun 10 · 21 min
No Priors: Artificial Intelligence | Technology | Startups
NVIDIA’s Jensen Huang on Reasoning Models, Robotics, and Refuting the “AI Bubble” Narrative
Jan 8
Books, tools, and gear mentioned in this episode
SignalCast may earn commission on purchases via these links. As an Amazon Associate, SignalCast earns from qualifying purchases.
Tools
“Running the Qwen 3.5 model locally means no API tokens or cloud costs, making autonomous robot operation viable for small businesses under $1,000.”
“Installing OpenClaw locally on a Jetson device enables natural language robot control — typing 'move arm up' or 'pick up object' executes physical commands without writing code.”
by NVIDIA
“Developers can mirror Seeed robot models inside NVIDIA Isaac Sim to validate and tune robot behavior before physical deployment.”
Gear
- SO-ARM robot armBy guest
by Seeed Studio
“Seeed Studio's SO-ARM robot arm costs $200 and ships ready to train without programming knowledge.”
by NVIDIA
“Users now train robot arms the way they train animals — physically demonstrating a task repeatedly, uploading the recorded data for cloud-based model training, then deploying via Jetson.”
More from NVIDIA AI Podcast
We summarize every new episode. Want them in your inbox?
Inside Instacart's AI-Powered Smart Shopping Cart | NVIDIA AI Podcast Ep. 302
How Mistral Is Building Frontier AI for the Enterprise | NVIDIA AI Podcast Ep. 301
Inside AI Tokenomics: How to Profitably Turn Tokens Into Business Value | NVIDIA AI Podcast Ep. 299
Snap’s Secret to Processing 10 Petabytes a Day: GPU-Accelerated Spark | NVIDIA AI Podcast Ep. 298
Harrison Chase of LangChain on Deep Agents, LangSmith, and Earning Trust | NVIDIA AI Podcast Ep. 297
Similar Episodes
Related episodes from other podcasts
Accidental Tech Podcast
Apr 9
686: Write Two Letters
No Priors: Artificial Intelligence | Technology | Startups
Jan 8
NVIDIA’s Jensen Huang on Reasoning Models, Robotics, and Refuting the “AI Bubble” Narrative
Lenny's Podcast
Aug 16
OpenAI’s Head of Design: This is the best time in history to be a designer | Ian Silber
Software Engineering Daily
Jul 14
AURA and Open-Source Agents for Production Operations
Cognitive Revolution
Jun 20
Dean Ball, on Joining OpenAI: New Power Centers, Frontier AI Policy, & Main Character Energy
Explore Related Topics
This podcast is featured in Best AI Podcasts (2026) — ranked and reviewed with AI summaries.
Read this week's Startups & Product Podcast Insights — cross-podcast analysis updated weekly.
You're clearly into NVIDIA AI Podcast.
Every Monday, we deliver AI summaries of the latest episodes from NVIDIA AI Podcast and 192+ other podcasts. Free for one show.
Start My Monday DigestNo credit card · Unsubscribe anytime