Michelle Wang

executive assistant @sundayrobotics

[Perry Jia @perryzjia]

1/N Two years ago, we recruited our first Memory Developer off Craigslist and onboarded them in a public library.

Today, more than 1,000 Memory Developers have helped us build the data engine behind ACT-2.

Here's how we got from that library to here 🧵

[Cheng Chi @chichengcc]

We just hit a weird milestone: our model became more reliable than your average home WiFi.

Just like everybody else, we thought cloud inference was the obvious choice. Yet 2 days into the ACT-2 eval, our mind completely changed.

If our hero @ArpitKalla didn’t cook, this video wouldn’t exist 🧵

[Michelle Wang @michlxwang]

@kaichan415 🍰🍰🍰

[Michelle Wang @michlxwang]

@johnson_zhsh new onboarding workflow

[jackie of all trades @looeegee_]

he's inevitable

[Michelle Wang @michlxwang]

@ChristineVan415 @tonyzzhao @chichengcc no you christine!! 🫶

[Michelle Wang @michlxwang]

one of the most hardworking teams I’ve had the chance to support!

over my last 4 months at Sunday, I’ve witnessed everyone come together despite chaos, last minute changes and manyyy long nights to put this update together. shoutout to my co-founders @tonyzzhao+ @chichengcc 🚀🚀

[Tony Zhao @tonyzzhao]

Introducing ACT-2 Preview

The first robotics model to unify broad generalization with high reliability. A single fine-tuning example can teach Memo a new behavior that generalizes.

Zero shot, real unseen homes, 99% success rate.

[Michelle Wang @michlxwang]

@bosonwang_ shots never miss😤😤😤

[Tony Zhao @tonyzzhao]

This fall, ACT-2’s first Solve enters homes through our Beta Program, the final step towards fully autonomous home robot deployment.

Full technical report: https://t.co/oVR2TAdlHg

[Michelle Wang @michlxwang]

vague-posting no more

[Sunday @sundayrobotics]

Introducing ACT-2 Preview, the world’s first robotics model that works in your home.

99% success rate, fully autonomous in unseen homes. Zero data from you.

[Sunday @sundayrobotics]

Introducing ACT-2 Preview, the world’s first robotics model that works in your home.

99% success rate, fully autonomous in unseen homes. Zero data from you.

[Michelle Wang @michlxwang]

the pile of laundry currently sitting on my couch is calling memo's name

[Tony Zhao @tonyzzhao]

We completed the most rigorous generalization test to date. Across 785 trials in 31 unseen environments, ACT-2 achieved 99.1% success in laundry folding.

ACT-2 also achieved human-level fold quality: receiving an average rating of 4.72/5, with 98.3% earning four or five stars.