Joined January 2021
- Tweets 74
- Following 182
- Followers 762
- Likes 264
3 Photos and videos
May 28
Such a great task choice @kylieyying ! Teaching robots to be confident in their movements and promote fair allocation of limited resources š
May 28
In my first week at @GeneralistAI, I trained a robot to pour liquids using GEN-1 š¤š§
I wanted to challenge the robot with a non-rigid manipulation task, so liquid felt like the perfect choice. The task involved:
- unscrewing the bottle cap
- pouring liquid into espresso glasses
- rebalancing uneven pours
Best of all, the robot was able to complete the task fully autonomously 3 times in a row (out of 3)! Pour-fect š
Excited for the journey ahead and grateful to be building alongside such an incredible team!
1
1
12
2,815
Coline Devin retweeted
Apr 30
Gen-1 ties zipties
Read more about Gen-1 in our blog posts in the comments below ā
6
28
217
36,135
Coline Devin retweeted
Apr 27
Today marks the end of my first full week @GeneralistAI
Last Monday, I was given a challenge: use our GEN-1 model to teach a robot a task of my choosing, using the same no-code platform our customers use.
I picked the ball-and-vase magic trick. It was one of my favorites as a kid, and it felt like the right mix of fun and surprisingly hard.
A few days later, GEN-1 pulled it off. I left Friday having watched the robot nail it 14 times in a row. Whatās wild is that even 4 months ago, if you told me you could go from idea to on-robot skill in a couple of days, I probably wouldnāt have believed you.
Really excited to be building with an incredible team. Canāt wait to see what week two brings š¤
14
31
319
38,930
Coline Devin retweeted
Apr 23
GEN-1 cleans white board
Read more about GEN-1 in our blog post in the comments below ā
7
17
245
62,972
Coline Devin retweeted
Apr 21
This was the first take for āI bet it would work with an arbitrary objectā š
Apr 21
GEN-1 plays the š shell game, trained on just 1 hr of robot data. It also generalizes to unseen objects, like @BerkayAntmen 's car keys.
Physical AI models should be capable of benchmark tasks like this one. It's interesting for the all the reasons @RhodaAI calls out -- requires visual memory, and the model must track the cups from the very start, at high frame rates.
Interestingly, GEN-1 appears to exhibit a degree of "active perception." It's subtle; the hands can sometimes appear to "follow" the cups, using its own movements to help attend to where it thinks the object should be.
Read more about GEN-1 in our blog post in the comments below ā
1
14
2,526
Apr 21
Memory is fundamental to robot intelligence. How can you know what to do next if you don't know what happened before?
Apr 21
GEN-1 plays the š shell game, trained on just 1 hr of robot data. It also generalizes to unseen objects, like @BerkayAntmen 's car keys.
Physical AI models should be capable of benchmark tasks like this one. It's interesting for the all the reasons @RhodaAI calls out -- requires visual memory, and the model must track the cups from the very start, at high frame rates.
Interestingly, GEN-1 appears to exhibit a degree of "active perception." It's subtle; the hands can sometimes appear to "follow" the cups, using its own movements to help attend to where it thinks the object should be.
Read more about GEN-1 in our blog post in the comments below ā
4
22
3,369
Apr 20
Our robots have been busy lately!
Apr 20
Everyday for the past 2 weeks, we've been sharing something new from GEN-1, our latest milestone in scaling robot learning. This has never been done before.
Going from ideas to skills in days (or faster) is what physical AI models should deliver.
More coming. Stay tuned.
Read more about it in our blog post in the comments below ā
2
27
3,987
Coline Devin retweeted
Apr 15
GEN-1 plays with fidget toy.
Other tasks GEN-1 can do:
youtube.com/playlist?list=PLā¦
Read more about GEN-1, our latest foundation model for the physical world:
generalistai.com/blog/apr-02ā¦
6
21
133
17,812
Apr 14
I love the little clicks when the pins are placed into the bowl <3
Apr 14
GEN-1 removes thumbtacks and papers from corkboard.
Other tasks GEN-1 can do:
youtube.com/playlist?list=PLā¦
Read more about GEN-1, our latest foundation model for the physical world:
generalistai.com/blog/apr-02ā¦
1
1
16
1,448
Coline Devin retweeted
Apr 13
GEN-1 puts plushies into polybags, in a warehouse outside the lab in New Hampshire.
14
60
441
102,094
Apr 10
An orange a day keeps the ...robot repairman away?
Apr 10
Orange you glad our robot can stack?
Other tasks GEN-1 can do:
youtube.com/playlist?list=PLā¦
Read more about GEN-1, our latest foundation model for the physical world:
generalistai.com/blog/apr-02ā¦
1
10
1,340
Coline Devin retweeted
Apr 9
I love this one, because deploying a robot then āintegrating with customer APIsā is not fun.
But a lot simpler when your robot can push the same buttons as a human šor use a touch screen.
Also, GEN-1 has memory too ā it learns to ārememberā which sock it put away, then taps.
Apr 9
GEN-1 sorts socks and uses a touchscreen stylus.
Done in collaboration with a partner on the Generalist platform. More on this soon.
5
8
58
8,805
You shouldn't need a PhD to teach a robot new tasks. Using our platform, partners can do this without a writing a single line of code.
Apr 9
GEN-1 sorts socks and uses a touchscreen stylus.
Done in collaboration with a partner on the Generalist platform. More on this soon.
2
2
45
5,527
Zip-zip
Apr 8
GEN-1 puts pens into zipper bag.
Robotās view, pixels into the model.
12
1,211
Coline Devin retweeted
Apr 8
GEN-1 puts pens into zipper bag.
Robotās view, pixels into the model.
6
12
181
27,404
Coline Devin retweeted
Apr 7
As a PhD student, I worked on methods to generalize across motion, objects, lighting, backgrounds, etc. Each is hard, and thereās always a long tail. (E.g. my robot still canāt handle new camera viewsāhence the āDonāt move! Thx :)ā sign.)
But if you have enough data to train from scratch, many of these generalization problems start to disappear. You get them almost āfor free.ā
Whatās powerful is the clarity you get when you start from the goal, not from a favored method: data is a key bottleneck to remove on the path to physical AGI. And working backwards from first principles, there may be ways to solve the data problem that arenāt actually that expensive.
That unlocks the freedom to train from scratch with simple methods, not crutches.
Apr 7
goals are more powerful than methods
5
55
4,366
Coline Devin retweeted
Apr 6
Generalist's new physical robotics AI brings āproduction-levelā success rates arstechnica.com/ai/2026/04/gā¦
1
4
6
5,022
don't need a robot to teach a robot
Apr 6
0 robot data.
Less than 2hrs of human data. Model opens its eyes and suddenly sees itās become a robot.
1
2
20
2,871
Coline Devin retweeted
Apr 6
This is GEN-1 putting paper bills into a wallet.
Paper has always been deceptively hard for robots.
Thin, deformable, and unforgivingāit bends, folds, and slips.
Not precise, you miss. Too much force, you crumple.
Easy for humans. But for robots, itās a full-stack challenge.
37
117
745
219,086
Coline Devin retweeted
Apr 2
Hear more about GEN-1 and improvisational intelligence from our own researchers @coolboi95 and @felixwyw
6
12
94
20,498