Co-CEO @yutori_ai. yutori.com. Previously: Senior Director, GenAI & FAIR at Meta, Associate Professor at Georgia Tech.
San Francisco, CA
Based in United States
- Tweets 4,695
- Following 154
- Followers 27,765
- Likes 7,874
Congrats @Jerr_Wu and team! 🚀
Today we're excited to announce @HalluminateAI's $30M Series A led by @oakhcft with participation from @YCombinator, @orangecollectv, @FTPartners, @heavybit, and more.
1
17
4,337
I haven’t been crocheting as much. And most of it has been very satisfying discs of varying sizes and colors. But this one I am excited about!
1
11
1,028
Computer-use in 2026
"What's left? The obvious answer is longer tasks. More steps, more hours. An agent that clicks for longer.
I think this is the wrong answer, ..."
"They can code. They can click. They can't choose."
"Choosing has to be trained, and it can only be trained where it matters."
4
11
2,215
n2 can drive your Mac!
Use Yutori's MCP to run our latest Navigator computer-use model locally.
Here it is in iMovie, cutting a quick video recap from our recent event.
1
5
28
4,242
Horsepower helped us communicate the value add of a new technology (steam engines) relative to a familiar status quo (workhorses).
What's that unit of measurement for agents?
Excited to have my Mousepower talk live on @aiDotEngineer's YT channel (link below), but I have retrospective thoughts on this topic & here is what I'd change if I gave the talk again today:
1. Thesis wasn't stated simply enough: value of agents is limited by our ability to measure them, otherwise we can't justify the ROI of their token cost
2. Need an explicit connection to the discourse on verifiable domains, which we need not be limited to today's set as understanding customer mental models unlocks new ways to verify value (e.g. horsepower)
3. After reading a nice post by @jon_stokes on verifiable domains, I'm specifically interested in the n-order F/X of their "optimization pressure"
4. Also rambled too much, perhaps a function of the topic being too underbaked... will see if GPT voice can coach me for next time
Follow up incoming (someday).
1
1
5
1,035
Most recent experiments
- WA-CHAA! peanuts
- Various protein snacks from IG ads
Current favorites
- Spicy Pizza savory protein bar from fellofoods.com + hummus
- Andy Capp's Hot Fries
- Cheddar Jalapeno Cheetos
- Garden Salsa SunChips
All-time favorite: Uncle Chipps Spicy Treat
2
7
811
I asked @Muse to scan my email and tell me what it has learnt about me. One of the things it concluded was I am a snacks enthusiast. 💯 ✅
3
49
14,520
A friend introduced me to junk journaling earlier this year. Turns out, I really like sticking flat things onto other flat things!
2
25
1,207
Neat to be on the Top 100 Women in AI 2026 list and part of this picture along with so many familiar faces from different phases and slices of my professional life :)
🙏
womenaibuilders.org/top-100-…
P. S. Took me a while to spot me :)
6
2
77
11,598
Highly recommend trying out nendo.world/.
I got this one made to see what I think of the quality, and I love it! I have now ordered several more. Several of my friends have also ordered a few.
Think your own art, your children's drawings, figurines of your loved ones doing their favorite things.
So much potential for delight!
2
3
13
5,197
This was wonderful :) Thank you all for coming! 🩵
An evening in SF with food, drinks, and conversation about all things CUA (computer-use agents).
Research, models, data, RL environments, inference, sandboxes.
With researchers and builders from frontier labs, big tech, fast growing startups.
Thursday.
Come join us!
@yutori_ai is co-hosting with @togethercompute and @CrusoeAI 💙
luma.com/aqmwya3d
1
3
33
5,151
Check it out!
The @yutori_ai website now offers an abundant garden of autonomous computers so you can test drive our new CUA model, Navigator n2, for free w/o an account. 🌻
1
16
2,773
🩵
Before each @yutori_ai launch, there's often a nice moment where engineers crowd my desk to see the final design touches. This one was specifically memorable, as our new n2 CUA model randomly awoke on its MacMini to start running autonomous tests next to us.
1
1
22
10,763
We’ve benchmarked n2 on long-horizon, real-world, computer-use tasks that require driving graphical and command-line interfaces across Linux and MacOS.
More details in our blog post.
yutori.com/blog/introducing-…
1
1
25
5,166