I love my Quilt(s). This will make them even better!
We are launching Quilt Tools, a permanent home for partner-facing tooling featuring the most advanced remote diagnostics platform ever made for residential HVAC. I talk internally about Quilt’s 3 Superpowers and this launch today ties all 3 together beautifully:
*Superpower 1, lt system has 10x as many sensors and 500x more computing power than the closest competitor. That data, that only Quilt has, and the ability to do sophisticated computation on it like system performance prediction and thermal modelling, is what makes this possible. It’s how we predict precisely when a system is going to break or tell an installer that they accidentally crossed wires or that they oversized an install. And the data is bidirectional – not only can our partners remotely understand the system, they often can remotely *fix* it as well. You’ve seen this capability in our OTAs that increase system capacity. This launch greatly expands those capabilities. The result is faster installations, fewer callbacks, and more comfortable homes – which saves money for both homeowners and partners.
*Superpower 2, Product*. Quilt is at its core a product company. We care a *lot* about great user-centric design, thoughtful execution, and quality. Quilt is staffed by people who have built some of the most used and loved software platforms in the world. We have taken all that experience to create a top tier enterprise software platform in a space that had never received this level of TLC. Wait, did I just say we are launching an enterprise SaaS platform in 2026?? Yes – because what makes it powerful and unique is our fleet of connected systems and the data it produces, which no amount of token spend can ever reproduce. Coding Agents have greatly *deepened* our ability to produce a great product with unique and defensible value here, not reduced it.
*Superpower 3, Partner Network*. Our channel partners and distribution network are our third superpower. Through them we have thousands of people in the field selling and installing Quilt every day. These are many of the best home services companies in North America. They are the face of Quilt and we hold just as high of a bar for partner performance and success as we do for our homeowner experience and products. And that’s 100% the focus of this launch – give our *partners* superpowers. We only succeed when they succeed and we want to give them every possible advantage to delight customers, grow their business, and accelerate heat pump adoption.
Big thanks to all our design partners who provided early feedback and are already using Quilt Tools in their daily work – this is for you. Hear from one of them, ARS Blueflame in Seattle, in my first reply below.
1
0
3
789
Michael Litt retweeted
Around 15 years ago, Kaheer Suleman and I started Maluuba with the vision of building Universal Turing Machines. Back then, we were one of the pioneer research labs building foundational modules of deep learning and Reinforcement learning with Turing award winners @Yoshua_Bengio ere was no AI hype then, as there is now. Those were the good old days of doing pure scientific research unlike the n+1 research that we see today. Maluuba was later acquired by Microsoft and became @MSFTResearch Canada.
Grounded in that same mission, we're finally ready to introduce @skyfallai to the world after a lot of experimentation in stealth over the last year. We're a frontier neo lab building the first Autonomous Enterprise by moving beyond the current LLM paradigm.
For the last 5 years, the foundational model market has relied on a single paradigm: scaling laws for LLMs - more data, more compute, bigger models. However, the real world is messy and much more complicated. In order to achieve our long term vision, the next generation of AI models requires a different approach. Skyfall is solving the hardest open problems in frontier AI: long-horizon planning, data inefficiency, and brittle performance in dynamic real-world environments.
To achieve the team's vision of a completely autonomous enterprise, the team is developing a next-generation frontier model (Enterprise World Models) using Continual Learning and World Modeling. Enterprise World Models can simulate the multi-layered consequences of strategic business actions. Our approach unlocks a new category in the foundation model market. To prove it, we're introducing Morpheus, a Continual Reinforcement Learning platform for AI researchers.
I'm building this company with the people I trust the most: my longtime friend Kaheer Suleman (prev. Co-Founder of Maluuba) and my brother @omgiamgod (prev. YC founder). Sumit and Kaheer are the first principles thinkers I can trust to go to the end of the world with to achieve the mission impossible together. Together with a stellar team of 25 researchers and engineers, we're pushing a new frontier in AI forward.
We unpacked our long term vision in today’s Forbes feature 🔗- read it to see what we’re building toward. Thank you so much Victor Dey for the interview.
To achieve our goal of enterprise world models, we are soliciting bids to acquire small SaaS startups (up to $1M) and fully automate them. If you’re interested, submit your business here: skyfall.ai/acquisition
Finally, thank you to our investors and advisors for believing in our vision since day one: @Fidelity , @sk121 (@touringcapital), @karam_n and @chrisarsenault (@inovia), @morgan_blumberg (@M13Company ), @stephpalmeri (@NextViewVC ), and @michaellitt and @mmccauley (@GarageCapital ), @jennydhe, @fchollet @NaveenGRao and so many others for supporting us in this journey.
Grounded in that same mission, we're finally ready to introduce @skyfallai to the world after a lot of experimentation in stealth over the last year. We're a frontier neo lab building the first Autonomous Enterprise by moving beyond the current LLM paradigm.
For the last 5 years, the foundational model market has relied on a single paradigm: scaling laws for LLMs - more data, more compute, bigger models. However, the real world is messy and much more complicated. In order to achieve our long term vision, the next generation of AI models requires a different approach. Skyfall is solving the hardest open problems in frontier AI: long-horizon planning, data inefficiency, and brittle performance in dynamic real-world environments.
To achieve the team's vision of a completely autonomous enterprise, the team is developing a next-generation frontier model (Enterprise World Models) using Continual Learning and World Modeling. Enterprise World Models can simulate the multi-layered consequences of strategic business actions. Our approach unlocks a new category in the foundation model market. To prove it, we're introducing Morpheus, a Continual Reinforcement Learning platform for AI researchers.
I'm building this company with the people I trust the most: my longtime friend Kaheer Suleman (prev. Co-Founder of Maluuba) and my brother @omgiamgod (prev. YC founder). Sumit and Kaheer are the first principles thinkers I can trust to go to the end of the world with to achieve the mission impossible together. Together with a stellar team of 25 researchers and engineers, we're pushing a new frontier in AI forward.
We unpacked our long term vision in today’s Forbes feature 🔗- read it to see what we’re building toward. Thank you so much Victor Dey for the interview.
To achieve our goal of enterprise world models, we are soliciting bids to acquire small SaaS startups (up to $1M) and fully automate them. If you’re interested, submit your business here: skyfall.ai/acquisition
Finally, thank you to our investors and advisors for believing in our vision since day one: @Fidelity , @sk121 (@touringcapital), @karam_n and @chrisarsenault (@inovia), @morgan_blumberg (@M13Company ), @stephpalmeri (@NextViewVC ), and @michaellitt and @mmccauley (@GarageCapital ), @jennydhe, @fchollet @NaveenGRao and so many others for supporting us in this journey.
29
28
94
72,263
/1 Excited to back @skyfallai's ambition: build Enterprise World Models to power the world's first autonomous enterprise.
The way I think about it is the "closed loop company".
Most companies today run open loops. You make a decision, then rarely find out if it was right.
The way I think about it is the "closed loop company".
Most companies today run open loops. You make a decision, then rarely find out if it was right.
4
1
9
589
@skyfallai /2 @skyfallai is closing the loop.
ie. A customer request becomes a spec, becomes product, becomes sold (or not sold) product, and the signal feeds straight back into the next decision, weighed against resources.
Now run this across every loop. Full operational context.
ie. A customer request becomes a spec, becomes product, becomes sold (or not sold) product, and the signal feeds straight back into the next decision, weighed against resources.
Now run this across every loop. Full operational context.
0
0
2
131
Michael Litt retweeted
All of my closest friends are now entrepreneurs.
Do what Harley is saying.
It's immeasurably worth it.
Some of the fine people in this photo - @danmartell @FitCoachCurtis @JaysonGaignard @michaellitt @darryltungsten
Do what Harley is saying.
It's immeasurably worth it.
Some of the fine people in this photo - @danmartell @FitCoachCurtis @JaysonGaignard @michaellitt @darryltungsten
Entrepreneurship is lonely. Especially early on. Nobody around you really gets it.
My advice: find other founders. People who understand the shared suffering of building something from nothing. They'll become some of the most important relationships in your life.
10
1
55
8,171
Michael Litt retweeted
Make Canada too business friendly to ignore. Especially for our own most capable.
Canada's startup exodus is accelerating. Nearly one of every two Canadian founders who raised over $1 million in 2024 are now based in the U.S. That's up from about one in five just a short time ago.
88
135
1,508
226,271
We’re back on the menu! 🇨🇦
After hearing feedback from Canadian founders in our network, we’ve decided to add Canada back to our list of accepted countries of incorporation.
Going forward, YC will once again invest in US, Canada, Cayman, and Singapore corporations.
ycombinator.com/blog/adding-ca…
2
0
3
519
Michael Litt retweeted
Today, we’re releasing a research preview of our real-time, local-first world model built for interactive, playable AI-worlds
60fps, locally run, all on consumer-grade hardware.
Come take a look ⬇️
60fps, locally run, all on consumer-grade hardware.
Come take a look ⬇️
54
147
1,090
497,992
Michael Litt retweeted
New world, 🌐 new year,⚡️new model! So much has gone into this wild pipeline - available as private beta. Waypoint-Medium is here! Join in the links below for access. #realtime #worldmodels #gaming
9
30
219
64,238
This is in no doubt an exciting moment for @GarageCapital, but it has to be said that we wouldn’t have been involved if it wasn’t for @mmccauley’s tenacity, conviction and heads down approach to pursuing great opportunities and staying on top of them.
Huge congrats @JonathanRoss321, @sundeep, & the entire @GroqInc team!
Jonathan, it was clear from our humble days at Google X that you had big ambitions. So happy for you!
Thank you for letting us @GarageCapital participate in the first round (and several since). What a ride 🙏
4
3
56
8,872
LSS, this investment was 15-20 years in the making and with @devongall’s support, we built a large position for our LPs (who got to meet Jonathan during our 2024 AGM).
Investors are a byproduct of their career learnings.
Investors are a byproduct of their career learnings.
1
0
10
468
@GarageCapital is a GP comprised of 3 @UWaterloo engineers turned founders, turned investors.
Proud to have leveraged our experience to find a seat at the table w the @GroqInc team and inspiring cast of co-investors.
Excited to apply what we’ve learned to our portfolio! ✌️🤙
Proud to have leveraged our experience to find a seat at the table w the @GroqInc team and inspiring cast of co-investors.
Excited to apply what we’ve learned to our portfolio! ✌️🤙
1
1
32
690
Michael Litt retweeted
Huge congrats @JonathanRoss321, @sundeep, & the entire @GroqInc team!
Jonathan, it was clear from our humble days at Google X that you had big ambitions. So happy for you!
Thank you for letting us @GarageCapital participate in the first round (and several since). What a ride 🙏
Jonathan, it was clear from our humble days at Google X that you had big ambitions. So happy for you!
Thank you for letting us @GarageCapital participate in the first round (and several since). What a ride 🙏
Today Groq entered into a non-exclusive licensing agreement with Nvidia for Groq’s inference technology. Along with other members of the Groq team, I’ll be joining Nvidia to help integrate the licensed technology. GroqCloud will continue to operate without interruption.
Learn more here: groq.com/newsroom/groq-…
29
4
151
50,585
Michael Litt retweeted
The Groq - Waterloo/Canada connection runs deep!🇨🇦
@mmccauley, Waterloo Tron '11 and @GarageCapital invested in Groq at seed. @chamath, Waterloo EE '99, did @groq's entire Series A.
Adrian Mendes, Waterloo EE '99, was Groq's first COO, for 7 years. @sundeep, Groq's current COO/President, is from UOttawa.
...not to mention the army of Waterloo/Canadians working in different engineering roles, including interns like @XanderChin, who demoed OpenGhost at @akatoshouse Fall Showcase.
Congrats to everyone, as a frequent Groq API user, $20 billion honestly seems too cheap. Nvidia got an absolute steal!
@mmccauley, Waterloo Tron '11 and @GarageCapital invested in Groq at seed. @chamath, Waterloo EE '99, did @groq's entire Series A.
Adrian Mendes, Waterloo EE '99, was Groq's first COO, for 7 years. @sundeep, Groq's current COO/President, is from UOttawa.
...not to mention the army of Waterloo/Canadians working in different engineering roles, including interns like @XanderChin, who demoed OpenGhost at @akatoshouse Fall Showcase.
Congrats to everyone, as a frequent Groq API user, $20 billion honestly seems too cheap. Nvidia got an absolute steal!
Nvidia acquiring Groq for $20B is a big deal.
Groq chips are insanely fast at inference, sometimes 10x GPUs. The trick is to put model weights in SRAM instead of HBM to trade memory capacity for speed.
Its $/token may lose to GPUs, but for long-wait inference on models like ers.
If I were Sam Altman, I’d buy it. $20B is cheap.
Jensen made the right call.
7
16
161
29,028