Recovering CEO (Wattpad), 3x founder (one homerun, one single, one strikeout), deep tech VC, electrical engineer, helping founders build world-class companies.
Canada 🇨🇦
Joined August 2007
- Tweets6.3K
- Following325
- Followers10.1K
- Likes6.5K
Pinned Tweet
Super excited to announce TSF’s Fund III today! With $41 million (US$30 million) to invest in the next frontier of computing and its applications, covering everything from deep-tech AI to protocols to semiconductors.
With a bigger fund, we will lead, co-lead, and participate in rounds, i.e., anything with the word ‘seed’ in it. In other words, from company inception to the round before Series A.
Our investment team members are all engineers, entrepreneurs, and operators, including a successful CEO-turned-investor—that’s me! :-) Our track record includes extensive experience in semiconductors, creating groundbreaking AI products, and founding and scaling massively successful global tech companies. Needless to say, we are uniquely qualified.
We will help your startup achieve extraordinary success, just like we did in the past. Please reach out to us if you are a seed-stage startup focusing on the next frontier of computing. We want to hear from you as long as you live on planet Earth!
betakit.com/two-small-fish-v…
Building a startup is a team sport of both IQ and EQ.
Brilliant technology alone won't build a great company. As CEO, your job is to set direction, bring extraordinary people together, and create the conditions for them to succeed.
A breakthrough technology is not a company.
Deep tech founders must master three things: technology, product, and commercialization.
And keep reinventing themselves as the company grows. The founder you are today shouldn't be the founder you are two years from now.
1/5 Historically, one of the golden rules of VC has been not to invest in competing startups. But AI is challenging that convention as startups increasingly expand into each other's markets. Even Benchmark is reconsidering how to navigate potential conflicts.
ALT Two business professionals shaking hands across a table.
4/5 That level of involvement requires trust. Founders share challenges and sensitive information with us that go well beyond board materials.
If we invested in competing companies, a perceived conflict could discourage that openness, undermining the value we provide.
5/5 Our reputation is worth far more than any single investment opportunity.
If being deeply hands-on is our superpower, protecting founder trust is non-negotiable.
That's why we don't invest in competing startups.
Full article: allensthoughts.com/?p=6815
Deep tech operates on proof and evidence. Startups operate in uncertainty.
As a founder, you can't wait for perfect information. You have to make decisions, move forward, and learn along the way.
Sometimes, good enough is perfect.
Allen Lau 🇨🇦 retweeted
Today we are releasing JST-2, the world's first full-stack physics-native world foundation model.
Many world models learn in pixel/video space. We went back to first principles and built the entire model stack in 3D.
Generate a 3D scene in seconds and expand it in real time. Create characters with motion built in, all through one general architecture.
Allen Lau 🇨🇦 retweeted
ByteShape is a Startup Sponsor at #PyTorchCon 2026!
Come visit us at Booth #52 for fun interactive demos, conversations about efficient AI, and a special surprise!
📸 Take a photo and bring home a printed keepsake!
📍 San Jose Convention Center
📅 October 20–21
See you there!
Happy Thanksgiving long weekend to my Canadian friends! 🇨🇦
While some people spend their cottage weekends fishing, I apparently prefer building a system to watch the fish. 🐟
Quite a few people asked how FishCam works, so I made a 5-minute explainer.
In between: secure networking, streaming software and automation.
What started as a side project has become a real-world technology playground for me and Two Small Fish Ventures.
Nothing beats actually building something to discover what works.
🎥 Watch the 5-minute architecture explainer:
youtu.be/3-xPFU9td80
🐟 Watch FishCam:
fishcam.twosmallfish.vc/
P.S. Of course, AI helped me write all the code. Who writes software by hand anymore? 😉
At the NBER Economics of Artificial Intelligence conference in Toronto, I spoke on a panel about China and its robotics industry.
One point I made: describing China’s robotics advantage as “cheap hardware” is both an understatement and an overstatement.
Our portfolio company @AcrylicRobotics is a good example.
Its secret sauce isn’t a new robotic arm. It’s the intelligence and know-how that let a generic arm translate artistic intent into physical execution.
Commodity hardware can increase the velocity of application-specific robotics companies by freeing them to focus on the brain, the know-how and the customer problem.
China can commoditize more of the robot without commoditizing the job.
substack.com/note/c-35516555…
