←
Yeok222 posts
Yeok avatar

Yeok

@yeokch

Building autonomous factories in the US. ex-UCL x Google Deepmind student researcher, before that was deploying robots into Hong Kong and China.

◷ Joined June 2023▧ 50 Photos and videos
Yeok avatar
Yeok@yeokch·Oct 17

I've spent years in China building robots. I can tell you this playbook won't work: Pre-seed $: Invent the ultimate actuator. Seed $$: Figure out mass production. The hard part isn't inventing a beautiful topology. It's process engineering. The real delta between the US and China is the ability to execute the enormous amount of physical x mental work required to iterate thousands of times on product designs and manufacturing processes. Partly because much of the tacit knowledge / management systems has left the US, but also because of the sheer human labor required. This isn't something you can solve by buying traditional automation lines either. It requires: 1. Ideation driven by human intuition. Walking up and down the line, talking to assembly workers, and developing hypotheses about what and where improvements can be made. Raw human hours and attention spent understanding a specific manufacturing process. 2. Large batches of experiments across multiple variables. Coordinating experiments that can easily involve 50+ people, issuing instructions, and tracking the execution and results of each experiment. 3. Purpose-built test design to identify key metrics. For example, a change in gear hobbing profile might require accelerated aging tests on a dyno, while a change in bearing tolerance might require NVH testing. 4. Data + human intuition-driven analysis. Interpreting test results and deciding whether to implement a process change or run the next set of experiments. 5. Strict process control. Reissuing work instructions to the line and ensuring every process is followed to the letter. The best manufacturers in Asia, from Foxconn to TSMC, have built their competitive advantage on systematically turning thousands of physical experiments into better manufacturing processes. Their moat isn't just production scale. It's the organizational machinery that makes manufacturing learn and improve faster than anyone else. I don't think most people in SF clock this. Say you're manufacturing actuator gears in-house in the US. After you've made your first article, will you hire a team of 10 process engineers to coordinate the 50 technicians running concurrent experiments with designs, processes, and materials? To understand wear patterns across every pin in a cycloidal drive after 10,000 hours? To reduce backlash from 15 arcmin to 5, then to 1? Just to keep pace with China's progress over the next few years? Do you have the skill set and management systems in place to do this? That's why our approach isn't just to automate manufacturing. It's to automate the process of learning how to manufacture better. Automate simulation and design optimization. Automate assembly and physical testing. Automate the design and execution of process experiments. Feed real-world experimental data back into simulation to continuously improve sim-to-real accuracy. The goal is to turn what currently takes 50 engineers coordinating thousands of experiments into a closed-loop system that continuously learns and improves. Then the same robotic line mass-manufactures the product. PS: Coming back to actuator design, many actuator designs are already commoditized in China. The project in the video is open source, and CNC machining cost in China is just $27 (watch the video). I hope people understand this reality. I'll link more examples in the replies.

Yeok avatar
Yeok@yeokch·Nov 13

Core beliefs of taiwanese manufacturing industry is 1. Trust, loyalty and the unconditional courage to stand up for others i.e. brotherhood - Guanyu 關聖帝君 2. Universal compassion, mercy, and household peace (ohno Jensen my bro) - Guanyin 觀音菩薩 To consistently manufacture complex hardware product is to ask your supply chain to put their necks on the line for your success. To invest in refining the process that makes YOUR product better years in advance, to trust that you will not bully them out of the payment terms - so your profit propagate to them next year. To ride it through the thick and thin together. I have no idea who in Nvidia can suceed his relationship with TSMC, Foxconn and the hundreds of downstream providers. You can find someone to replace Jensen as a decision-maker but I'm not sure you can find someone to replace his presence as the head of the east-west supply chain for the entire semiconductor industry. And I freaking hate cheaters, I'm just telling you facts on manufacturing. Penalize him according to your moral codes yes but a CEO stepdown like this with no real successor and we will start seeing the microtears in the social fabric grow into rifts in a few years.

Post attachmentPost attachmentPost attachmentPost attachment
Yeok avatar
Yeok@yeokch·Nov 13

Guys maybe I should have specified what drive wth ahhahahah

Yeok avatar
Yeok@yeokch·Nov 13

I asked our Taiwanese factory’s owner why he’s sharing all their manufacturing secrets with me He said it’s because I’m a good looking young man with incredible drive Maybe the real honeypot was me all along

Yeok avatar
Yeok@yeokch·Nov 10

Need to rename event to “researcher or honeypot two truth one lie” Sadly @Chloeeewrx_ vets everyone which takes out 99% of the suspense 🙂‍↔️

For those of you DMing me, yes, we’re throwing another one at the house tomorrow, bringing together SF people in robotics, AI, and all things frontier tech! See you all soon :) luma.com/yeu7dx2f

Yeok avatar
Yeok@yeokch·Nov 10

Every QDD today like YAMs, Piper has a payload around 3kg RAIPAL: 20kg allegedly, probably closer to 10kg in real world tasks but still a huge leap 3k planetary gearbox on an inrunner btw

Post attachment

韓国KAISTが、自重の2.25倍の20kgを持ち上げ、先端を秒速20.6mで振れて、ハンマーで打たれても動き続けるヒューマノイド用の腕を開発しました。 人の腕の力、速さ、持久力の計測値から目標を決め、上腕は力、前腕は速さを重視した部品で組みました。肘は姿勢によって力と速さの配分が変わる仕組みです。 熱の解析では、ほぼすべての負荷で、人が同じ重さを持ち続けられる時間を上回りました。

1:58
Yeok avatar
Yeok@yeokch·Nov 8

this was when his ego took over :( i remember telling people about it but they're like nah let him have his superstar limelight moment no man. as a guy i know. his brain did some quick maths, he acted on it, and that rewired his brain and he was never quite the same.

$NVDA Jensen Huang signed the chest (the now infamous photo) when he was in Taiwan for the Computex expo in June 2024. Guess when the first public photo of supposed “Chinese Spy” Alice Huang and Jensen Huang was taken and where? June 2024 in Taiwan as well. Bro wanted all the smoke on that trip… lol

Post attachmentPost attachment
Yeok avatar
Yeok@yeokch·Nov 8

Something like this I believe - cycloidal-like but moved the meshing off the radial plane into an angle, which makes it more like a coin spinning and wobbling to the ground which the article calls precession but also not exactly? Some thinking / optimizer needed to make the constantly changing cog and inertial moment reasonable Or add a second disk but that’s kinda wild machining work Spinning watermelon gearbox by cracked Japanese engineers 🍉🐐 x.com/amesekka/status/186717…

球面差動減速機-1/63 クリスマスカラーで

0:10
Yeok avatar
Yeok@yeokch·Nov 8

patiently waiting for japanese engineers to figure out the next zero-backlash-non-planar-cycloidal 🥹 👉 👈

0:25

新機構の減速機、同軸ローラーカム 歳差運動でロボット精度を工作機械級に xtech.nikkei.com/atcl/nxt/co…

Yeok avatar
Yeok@yeokch·Nov 6

1 raghav + 1 yeok ~= 1 alan 🐐 unit of intensity (for now) Time to ramp like crazy 🚀

Was a full circle moment to introduce @alanchanguk to the team and show all the fast progress we’ve made all Caldera. I joined @fuseenergy when it was 40 people and worked closely with Alan, Charles as the company went vertical, from 20M to 600M ARR in 2 years. Early growth like that is intoxicating and visceral - you put in 80 hour weeks not because someone asks you to, but because you feel a strong sense of belonging to the business - you think if not you, then who? We’re bringing that ruthless pace to Caldera. At Fuse a lot of priors were set from Revolut from which it derived its culture. Now I catch myself saying, far too often, ‘This is how we did it at Fuse’. If you’re a killer engineer and want to rethink manufacturing, this is where you start.

Post attachment
Yeok avatar
Yeok@yeokch·Nov 4

Hosted the highest signal roboticist / hard tech house party in SF (allegedly) 🎉 Thanks for coming and biiig thanks ti cohosts @itspublu @Chloeeewrx_ @firstc0in @_tanvi_bansal @librendaa072 wohooo

Post attachment
0:22
0:04
0:09
Yeok avatar
Yeok@yeokch·Nov 3

Would host again 🙌

Inside the living rooms powering SF’s Physical AI boom. 🤖

0:38
Eric Jang avatar
⟳ Yeok retweeted
Eric Jang@ericjang11·Oct 31

He that is without sin among you, let him first accuse another robotics company of doing too many demos

Yeok avatar
Yeok@yeokch·Oct 29

I can see a world where this works I can also see a world where it doesn’t because trump doesn’t understand the manufacturing stack 50/50

Post attachment