emily yu
@emily_yu
Followers
5K
Following
3K
Media
52
Statuses
706
hardware, matcha, overthinker | investing @BoostVC | ex- @amazon robotics, @nfx | @mit | 🇺🇸
San Francisco, CA
Joined December 2010
writing $50k to $1M check to founders looking for no bs investors who move fast. life is too short, spend it on building not fundraising.
264
71
2K
2026 pivots: crypto -> robotics ai -> robotics manufacturing -> robotics agents -> robotics 3d printing -> robotics and if you want to sound really smart, just replace robotics with embodied ai
54
38
473
The @BoostVC Founder Summit kicks off! 21 companies Bio, Energy, AI, Industrials, Robotics, Space Founders from US, Canada, England, China, Norway, Italy, Ireland, Germany The future is going to be awesome.
2
5
64
if you have a laptop and internet, you have distribution and leverage. in 2026, the bottleneck isn’t credentials, capital, or connections. it’s ambition. build something unreasonable.
12
4
57
the unlock of dexterous manipulation, @DanielXieee & Origami is paving the way. gearboxes kill force transparency and break sim2real. rare to see a team with both the technical depth and the deployment mindset. the ones who think about real world constraints from day one are the
Why does manipulation lag so far behind locomotion? New post on one piece we don't talk about enough: The gearbox. The Gap You've probably seen those dancing humanoid robots from Chinese New Year. Locomotion isn't entirely solved; but clearly it's on a trajectory. But we haven't
1
1
25
the data problem for physical AI isn't volume, it's modality. tactile, contact-rich data doesn't exist at scale. cool to see @PalatialSim and @stevenRSim building the pipeline to generate the missing input layer for embodied intelligence
A child consumes more data in 1 month than any LLM has ever seen. Embodied agents learn by doing, but the data that teaches them is tactile, sensorial and causal. Such data does not exist. To make physical AGI possible, we need to generate this new data at an industrial scale.
6
5
32
these days i can't tell: - real vs. ai video - friends vs. their agent - flirting vs. a chatbot with good rizz - using a product vs. being the product maybe agi / asi is just a permanent state of confusion
4
1
12
when you put mit researchers, boston dynamics, uc berkeley, princeton, and ex-spacex founders around one robotics dinner table… some of the best takes from dinner last night: - tactile sensing is non-negotiable for dexterous manipulation, especially in unstructured environments
11
4
91
Radiant gained a new strategic partner: @LockheedMartin chose @RadiantNuclear to invest. July 4th, a new era of nuclear energy begins.
14
18
118
emily was a believer before most all others. technical investor with conviction. incredible respect and honored to call her a friend :) đź‘‘
everyone talks about taste. @jia_seed and @mal_shaik have been quietly building it for a while. creativity + obsession with detail + speed. one to watch.
7
1
30
a few years ago, my founder friends waited till series a to get a gf. now they’re waiting till series b.. guess valentines moved upmarket with the ai multiples.
6
0
35
disclosure: backed jia in jan 2025 after meeting her at an event on her 21st birthday. conviction was immediate.
1
0
9
everyone talks about taste. @jia_seed and @mal_shaik have been quietly building it for a while. creativity + obsession with detail + speed. one to watch.
3
1
31
you know you’ve been in the ai bubble too long when you see a kid on the plane doing word puzzle and your first thought is “why isn’t she claude coding?”
2
0
15
interesting vc dynamic in 2026: vc firms: go big, or go first. individual investors: go viral, or go technical. the middle is disappearing.
11
3
73