Dyna Robotics

General and robust AI robots that power the future of the physical economy.

[Dyna Robotics @DynaRobotics]

We are releasing Dyna-2.1, the first Physical Agent that achieves reliable super long-horizon whole-body autonomy. It combines our brand-new semi-humanoid hardware with an agentic system built around Dyna-2 to handle ultra-long real-world workflows.

Here is an uncut footage of Dyna-2.1 completing an entire hour-long laundry room workflow, just like a human does.

[Conference on Robot Learning @corl_conf]

🔷 Thanks again to Patron Sponsor @DynaRobotics! See you in Austin at #CoRL2026 🤠🤖

[Dyna Robotics @DynaRobotics]

You know the most interesting conversations happen outside the conference walls.

In Austin for CoRL? Join the Dyna crew and some of the brightest minds in robotics for an evening of food, drinks, and great conversations.

Space is limited, so grab your spot!

🗓️ Wednesday, Nov 11 · 6–9 PM CST

RSVP: https://t.co/I9CmVBMqaX

[Anh-Quan Pham @_anhquanpham]

not many things can demonstrate production grade quality in robotics models better than an unexpected instagram reel on a sunday morning

(this is not our only hotel deployment btw 👀 it was indeed the first iirc)

[Dyna Robotics @DynaRobotics]

Spotted on the job by a hotel guest 👀

Our robots take towel duty seriously. If you spot them too, tag us!

📹: travelreligiously on Instagram
https://t.co/71w8FXRlHj

[York Yang @YorkYang5050]

https://t.co/sVIlj3X99i

This is real!

[Jason Ma @JasonMa2020]

This is already the present for SOTA robotics architecture

[Anh-Quan Pham @_anhquanpham]

Growing up, my biggest dream was to become a professional soccer player performing at the highest level.

I never made it to the senior pros. Life had other plans, as they say. Somehow, a random kid growing up with the ball along the Mekong river in Vietnam ended up at Penn, where I got my hands on a robot for the first time. I spent nearly all my waking hours there working on reinforcement learning and robotics research, and eventually came to Dyna as a research scientist.

There's something about being a professional athlete that every kid dreams of. You wake up knowing what you want, spend an unreasonable amount of time getting better at it, and care about winning probably more than you should. It's that sense of purpose in what you do, and the feeling of chasing mastery.

The other day, I pushed this one training feature that shot up our models' precision and capabilities across our office experiments and deployments all at once. A coworker dropped a Messi photo in Slack (swipe for the receipts 😭). I was smiling at my screen for a while.

I think that soccer dream never really faded. I just found something else I wanted to master along the way. And at Dyna, I've found the other parts of what I loved about soccer: a locker room locked in on the same goal, and teammates who somehow make you want to run another kilometer when your legs are already completely dead. You look to your left and right, and some of the best players of this generation trust you with the ball just as much as you trust them.

​I guess robotics & AI is my sport now. In a way, even the models I train are out there competing in their own sport, where quality, throughput, and reliability across all capabilities are the goals they have to score, no matter what the real world throws at them. And ofc, the models (and Taku 🤖) are obsessed with learning and mastering all their crafts too!

It's been quite an adventure working on Dyna-2, Dyna-2.1, and bringing Taku to life. Pretty damn happy I get to keep making history with this team. We've got a lot more goals to score ⚽️🦾

[Conference on Robot Learning @corl_conf]

🔷 Thanks to @DynaRobotics for being a Patron Sponsor of #CoRL2026! 🤖
👋 See them in Austin

[Margaret Meagher Berberi @Margaret_Mar_]

Working at Dyna is the definition of team work. Cross functional development is core to how we build and develop, creating the perfect environment to refine your technical skill set.

Loving what you do is the norm here 🤖

[Prashant Srivastava @prash_srivastav]

What drew me to Dyna Robotics was the opportunity to solve software problems that go beyond software itself.

Having spent years building consumer robotics at Amazon, I knew firsthand how challenging it is to make robots work reliably in the real world.

[miles kane @lifeofmlk]

For the last decade, my craft has been 0 → 1 GTM.

Finding the right first customers and use cases. Defining what success looks like. Testing the motion until it repeats.

In a lot of ways, it all feels like it’s been building toward Dyna.

Here, that means getting out of the office and into the places where Taku will actually work.

On a recent trip, Amy, our Head of Product, Skyler, one of our deployment engineers, and I spent the day on the road deep in the East Bay. We walked four hotel laundry rooms where Taku could go to work next. We hadn’t even started a POC yet.

Every hotel was different. Towels, carts, layouts, workflows. Things you don’t learn on Zoom.

We’re doing a lot of this right now: getting into the field early, learning what changes from site to site, and bringing those lessons back to the team as we build a repeatable motion.

We each own a part, but we’re all responsible for whether the whole thing works.

Proud to keep sharpening my craft alongside this team. Here’s more about what that looks like at Dyna.

[Shubham @machal_shubham]

One thing I love about robotics is how much thought goes into the smallest details. Really cool to see that craftsmanship celebrated here.

[Kun Huang @kun_h____]

A friend at Dyna once told me his career goal is to put 100 million robots into the world. I think that sums up why many of us are here: an ambitious vision, backed by a grounded understanding of what it takes to get there.

Robotics is a complex systems problem. No matter how much you know, there's always another part of the stack you don't. What I love about Dyna is being surrounded by people who are masters of their own craft, and getting to learn from them every day.

Building useful robots takes much more than a great model. It takes the full stack, and a team that knows how to bring it all together. If this sounds like the kind of challenge you'd love to work on, come join us!

[Mohid Khan @Mohid169]

I joined Dyna because I wanted to work alongside people who genuinely love robotics and share the ambition of bringing robots into everyday life.

What excites me most is the opportunity to help turn decades of progress in robotics into machines that can meaningfully contribute to society.

It's a privilege to work with a team that shares that passion, and I'm excited to keep learning, building, and refining my craft alongside them.

[Cong Gao @CongGaoJHU]

Management efficiency at its best: a flat team, zero cross-functional friction, and the freedom to tackle problems that matter. Whenever we need resources, teammates are always eager to help. Check out our craft video below: 📷

Interested to apply? Check https://t.co/4FmOMxPDGV

[richard zhang @rzhang139]

My favorite moments are iterating with robots in-house, gathering UX feedback, and thinking through on-edge resource tradeoffs across observability, research, and deployments. Nothing is too big for an idea but the smallest, intentional contributions compound tremendously.

If you’re a fellow systems thinker, join us! https://t.co/B5CGxyvukT

[Yannis He @yannis__he]

From the audience seat, the robotics industry is always surrounded by celebration: the fancy demo, the massive raise, the spotlighted stage, all the fun stuff.

When you're in the game, you know those "dopamine moments" only last a few minutes.

It's the few minutes an athlete stands on the podium, after hundreds of thousands of reps.

It's the few minutes a musician plays one piece to a full hall, after years of rehearsal.

It's the few minutes a rocket climbs off the pad, after years of engineering.

It's the few minutes a craftsman hands over the finished piece, after a lifetime of refining the same cut.

If you only enjoy those few minutes, I'm afraid the robotics game will disappoint you. If you enjoy the craftsmanship behind them just as much, welcome to the game.

[Ethan Jahn @ethanwjahn]

Those who love walking will walk further than those who love the destination

[dan efroni @EfroniDan]

At DYNA, I get to work across hardware, perception, and control with teammates who challenge assumptions and work through hard problems together.

If building robots people can depend on sounds like your kind of work, come join us.

https://t.co/AA1M0xwTuz