The Peel

1.2K posts

The Peel banner
The Peel

The Peel

@ThePeelPod

Exploring the world’s greatest startup stories. hosted by @TurnerNovak. Watch full episodes 👉 https://t.co/ZRJGDoYMhV

Katılım Haziran 2023
5 Takip Edilen3K Takipçiler
Sabitlenmiş Tweet
The Peel
The Peel@ThePeelPod·
What @jimbelosic thinks was the hardest part about bootstrapping his manufacturing company @sendcutsend to $140M ARR: "The hardest part is doing many, many things at once. And you need to resist automating it for as long as possible. We were lucky because we started by just cutting sheet metal. A lot of people told us that was never going to work because sheet metal fabrication really needs a hundred more steps. Bending, painting, anodizing, whatever. I said, “Nope. I’m just going to cut it.” Then we slowly added over time. What I see now is people saying, “I’m going to build a factory and make bicycles.” It’s incredibly complex. You need the raw materials, cutting the tubes, welding, managing supply chain for gears and sprockets and pedals and all the little parts, then the assembly. Doing all of that at once is really challenging. It’s almost overwhelming. And what people try to do today is automate it all. And that’s where things go really sideways. You can over-automate. We’re very sensitive about when to use people and when to use robots. We put robots where robots make sense. But we also have great people who can adapt within seconds and move all across the factory to where they’re needed. Automation is very difficult to set up accurately the first time. Doing everything at once in manufacturing is hard. If you have the luxury of a slow ramp, start with one very simple product. If I were going to make a humanoid robot, I’d start with just winding copper around an armature. Then make it into an actuator. Then maybe make some PCBs. Then make the cameras. All over months and years, before I had to do the whole humanoid. People with much more ambition and resources are doing the whole thing at once, but it’s very, very hard. So start with something, do it manually, get really good at it, and then automate around the edges as you expand. And we were able to do this because we were bootstrapped. We had no pressure to deploy capital fast. And I think that's one of the biggest problems with venture right now. They're throwing a shit ton of money at people without a ton of experience, with the expectation that you figure it all out very quickly because they need to make their money back. And it takes longer than that. Where you start and where you end are often different. You might start out trying to make toasters and it turns out you’re really good at the electrical components, so you become an electrical component manufacturer instead."
English
0
2
5
1.7K
The Peel retweetledi
Ashwin Varma, MD
Ashwin Varma, MD@varma_ashwin97·
Moats aren't real. My #1 most contrarian belief.
Turner Novak 🍌🧢@TurnerNovak

From @jimbelosic on how @sendcutsend uses speed and quality as a competitive advantage against overseas manufacturers: " I don't know if there's any moat, anywhere. Because there's so many smart people that have figured stuff out. If you think you have a moat, that means you think that people being born today aren't gonna be smarter, or more driven, or have more energy than you. Like, it's gonna happen again. So with us, our moat is continuing to deliver and exceed expectations. And really, it's speed. A lot of speed is probably the best moat that we have. Because even if someone can sell it cheaper, if we can get it to them faster and they can complete their project faster, that's the most important thing. So speed, capacity, and quality. Think about it Chick-fil-A. You know, Popeye's sells a chicken sandwich. McDonald's has a chicken sandwich. But people love Chick-fil-A because it's high quality. It's consistent. Maybe the locations are convenient, the drive through's pretty fast, even though the line is long. So I don't know if that's a moat or not. But they're running a really good business. So we try and do the same thing. Internally, we always run the teleport model. Where we're like, okay, let's say China could teleport. And as soon as they put it into a box, it magically crosses that ocean, and then gets delivered. So all of a sudden, we have no speed moat. What does that look like with our business? And our answer to that is, it's the entire experience. It has to be the look and feel of the product. It has to be the support. Before the sale, after the sale, during the sale, or whatever. The tough thing about manufacturing is DFM. Designed for manufacturability. So just because you can design something, doesn't mean that it can be made. Especially to all of you people listening with 3D printers. Just because you can 3D print something, doesn't mean that it can be made on a CNC machine, or injection molded, or whatever. So you have to make compromises in your design, or adjustments in your design in order for it to be manufactured. We invest a significant amount of resources into helping our customers understand how to make their design better. How to make it cheaper. How to make it faster. And that's something that offshore has challenges with, is the support and partnership."

English
1
2
6
2.8K
The Peel retweetledi
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
.@jimbelosic on treating manufacturing like baking cookies to bootstrap to $140M ARR in eight years
Turner Novak 🍌🧢@TurnerNovak

New @thepeelpod with @jimbelosic Jim bootstrapped @sendcutsend to a $140 million revenue run rate in eight years. We talk building a sheet metal manufacturing business in the US, creative ways he financed the company early on, using speed, trust, and software to compete with overseas competitors, lessons from restaurants, and why you can’t run factories from a spreadsheet. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 0:16 Automating sheet metal manufacturing 5:59 Zero to $140 million ARR in 8 years 7:58 Acquiring a $750k laser with $0 13:38 Automating factories is like baking cookies 15:17 Being legible to capital 17:31 Unlocking custom, low order manufacturing with software 20:00 Building more factories instead of selling the software 24:50 Run your company like a lemonade stand 28:30 Raising an angel round in 2021 as a safety net 33:21 SendCutSend’s unique bottoms-up GTM 38:24 Fun coupons 40:12 Building a moat with speed and trust 45:55 How US factories can beat China 47:40 Gaslight product launches 52:05 Lessons from non-manufacturing businesses 55:19 You can’t run a factory from a spreadsheet 58:10 Using data in manufacturing 59:50 Lessons from Factorio 1:03:17 Unlocking a negative cash conversion cycle 1:06:14 You need to resist automating everything 1:13:51 Surviving COVID with six weeks of cash 1:15:47 Solving the US skilled labor shortage 1:26:17 Teaching kids about manufacturing

English
4
4
12
3.9K
The Peel retweetledi
Jim Belosic (SendCutSend)
This was a fun one. Timestamps below
Turner Novak 🍌🧢@TurnerNovak

New @thepeelpod with @jimbelosic Jim bootstrapped @sendcutsend to a $140 million revenue run rate in eight years. We talk building a sheet metal manufacturing business in the US, creative ways he financed the company early on, using speed, trust, and software to compete with overseas competitors, lessons from restaurants, and why you can’t run factories from a spreadsheet. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 0:16 Automating sheet metal manufacturing 5:59 Zero to $140 million ARR in 8 years 7:58 Acquiring a $750k laser with $0 13:38 Automating factories is like baking cookies 15:17 Being legible to capital 17:31 Unlocking custom, low order manufacturing with software 20:00 Building more factories instead of selling the software 24:50 Run your company like a lemonade stand 28:30 Raising an angel round in 2021 as a safety net 33:21 SendCutSend’s unique bottoms-up GTM 38:24 Fun coupons 40:12 Building a moat with speed and trust 45:55 How US factories can beat China 47:40 Gaslight product launches 52:05 Lessons from non-manufacturing businesses 55:19 You can’t run a factory from a spreadsheet 58:10 Using data in manufacturing 59:50 Lessons from Factorio 1:03:17 Unlocking a negative cash conversion cycle 1:06:14 You need to resist automating everything 1:13:51 Surviving COVID with six weeks of cash 1:15:47 Solving the US skilled labor shortage 1:26:17 Teaching kids about manufacturing

English
4
3
62
9.9K
The Peel retweetledi
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
New @thepeelpod with @jimbelosic Jim bootstrapped @sendcutsend to a $140 million revenue run rate in eight years. We talk building a sheet metal manufacturing business in the US, creative ways he financed the company early on, using speed, trust, and software to compete with overseas competitors, lessons from restaurants, and why you can’t run factories from a spreadsheet. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 0:16 Automating sheet metal manufacturing 5:59 Zero to $140 million ARR in 8 years 7:58 Acquiring a $750k laser with $0 13:38 Automating factories is like baking cookies 15:17 Being legible to capital 17:31 Unlocking custom, low order manufacturing with software 20:00 Building more factories instead of selling the software 24:50 Run your company like a lemonade stand 28:30 Raising an angel round in 2021 as a safety net 33:21 SendCutSend’s unique bottoms-up GTM 38:24 Fun coupons 40:12 Building a moat with speed and trust 45:55 How US factories can beat China 47:40 Gaslight product launches 52:05 Lessons from non-manufacturing businesses 55:19 You can’t run a factory from a spreadsheet 58:10 Using data in manufacturing 59:50 Lessons from Factorio 1:03:17 Unlocking a negative cash conversion cycle 1:06:14 You need to resist automating everything 1:13:51 Surviving COVID with six weeks of cash 1:15:47 Solving the US skilled labor shortage 1:26:17 Teaching kids about manufacturing
English
10
19
207
63.6K
The Peel
The Peel@ThePeelPod·
“We had unit economic fit. We had product market fit. But we didn’t have go-to-market fit.” @Alex__Israel on the growth buyout strategy @metropolisio used to scale from zero to the largest parking lot operator in the world in nine years.
English
1
2
5
3.4K
The Peel
The Peel@ThePeelPod·
Parking is 15% of cities surface area. @Alex__Israel explains why it never became an institutional asset class: "The cash flow is sticky. It's low maintenance and CapEx. But it's also unsexy. You can't really brag about owning parking lots like you can with other types of real estate. It's dirty. It's gritty. It's a piece of pavement. There are weird smells. It's the last bastion of non-institutional real estate in the US. When we started @metropolisio, we figured out we could create technology that created a better mousetrap: 1) Consumers could drive in and drive out. 2) This reduced the cost to operate the facilities 3) We could also charge more revenue It was a complicated sales cycle at first. But all of this increased the value of the underlying property, and now it sells itself."
English
2
3
8
3.5K
The Peel retweetledi
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
New @ThePeelPod with @Alex__Israel, co-founder and CEO of Metropolis. They're a nine year old startup that's also the largest operator of parking lots in the world. We unpack their venture growth buyout / AI rollup strategy that's raised billions to acquire legacy parking assets, and open up the playbook for other founders and investors. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 1:10 Helping 50m Americans park 4:00 Building “Buy Now” for the physical world 9:02 Real-world checkout technology that works 16:07 Why parking never institutionalized as an asset class 18:34 Using tech to make parking assets more valuable 21:53 Parking lots as autonomous robotics hubs 29:07 Going to film school, working at MTV 30:55 Starting his first parking data company 33:47 Culture of pranking each other 36:27 A Fortune 500 CEO convinced him to start a 2nd parking company 42:55 Realizing they couldn’t sell to real estate operators 46:09 Acquiring a company 10x their size to jumpstart GTM 50:20 How to do a successful AI growth buyout 54:48 Revenue growth must be driven by technology 1:00:33 Why companies should do growth buyouts 1:03:55 Being legible to capital 1:09:16 You need creativity to take risks 1:13:30 AI is the first ever disruption to skilled labor 1:19:14 CEO challenges growing zero to 23,000 employees 1:24:31 Alex’ personal AI stack
English
9
5
21
21.1K
The Peel
The Peel@ThePeelPod·
How does science change when you combine humans and AI agents together, all running completely decentralized? "MIT has developed something called ScienceClaw. It’s the next generation of collaborative science. We showed a couple of examples it found last week. The first one: what do cricket wings, baroque choral music, and composite materials have in common? We had agents in biology, agents that understand material properties, and agents that understand music, all collaborating in a decentralized way, finding a part of design space that was unexplored. It ended up discovering some incredibly useful material resonators. The operational use is very broad. It’s not like this was discovered today and I’m going to use it tomorrow. But if you can now design material resonators for any property you want, and agents are finding the process to achieve that design, that’s a real step forward. The second application we showed is superconductivity, which is more straightforward to understand. If you can pass energy across a medium with zero loss, no dissipation, that’s a superconductor. The problem is pretty much all known superconductors only work at extremely low temperature. Close to absolute zero, -100 to -200 degrees. Which is just not practical for most applications. We’re using agents to search for superconductors that work at much higher temperatures. Room temperature, ideally. I don’t want to give the impression that we ran the agent and won a Nobel Prize. This is the first step in a long chain. Next, you have to build the material, test it, get information, come back. But this first step is actually the hardest one. The possible space of superconducting materials and configurations is enormous. Without these newer techniques, searching that space would take a very, very long time. AI speeds up that first step in scientific discovery tremendously."
English
2
6
9
4.3K
The Peel retweetledi
Chris Hladczuk
Chris Hladczuk@chrishlad·
"Company values" are mostly generic and useless. So I wrote a manifesto titled "How to Succeed at @hanoverpark" instead. Every new joiner reads it on their first day. Here's one part about our values... 1) Egoless Execution A willingness to do the schlep is important at Hanover Park. The best people take work off your plate rather than piling it on. And being senior is no exception. All individuals need to: - do individual contributor work as their primary job. - develop the skills to ruthlessly prioritise tasks with the business goals in mind. - Be hungry for as much context as possible to make high quality high level decisions on your own 2) Maniacal Urgency No startup has ever built a multi-entity general ledger from scratch in 6 months or migrated billions in assets in months. Our entire company lives and dies on urgency. Working at Hanover Park is not the type of satisfaction you feel from relaxing on the beach. It’s the satisfaction you feel from the best workout class of your life. This level of urgency will be uncomfortable for you, our vendors and anyone we encounter. However, it’s the standard we must demand of ourselves to make our customers raving fans. This means… - We hate it when someone tries to schedule a meeting for next week instead of the same day. - We ruthlessly question timelines… “what would need to be true to accomplish this 1 month goal in 1 week?” 3) Own the Details Building financial infrastructure is all about the specifics. This means debugging complex allocations. Or wrangling messy historical data. You are owning the details if you physically traced the excel sheet migration file and know about the incorrect date format on row 27 based on the historical PDF capital call notice to debug a migration. The people who are not deep in the details and go to the source waste everyone’s time - this includes me as CEO. I will likely have less details than you on a topic - so I expect you to tell me I’m wrong. You earn the right to a strong opinion by owning the details. -- What do you think about these values?
English
3
4
27
10.4K
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
TBPN on Monday MTS Live on Tuesday Sourcery office tour on Wednesday 20VC in London on Thursday We are living in the golden age of going on podcasts
English
20
2
110
8.6K
Flex
Flex@FlexSuperApp·
Who’s based in SF and wants to be a full time shit poster?
English
16
0
34
3.1K
The Peel retweetledi
Chris Hladczuk
Chris Hladczuk@chrishlad·
How we're building our engineering team culture at @hanoverpark:
English
0
4
17
12.8K
The Peel
The Peel@ThePeelPod·
Last week researchers at the University of Michigan demoed two new scientific discoveries, made entirely with AI. I asked this professor how AI is transforming scientific discovery: "Let’s break down the scientific process. You make observations in nature. You write down a theoretical model that explains the observation. Then you manipulate that model to get the property you want. This is the scientific method that we all learned about in high school. Theory, observations, experiments. Then you go to more detailed models, the kind that a lot of people in the institute I direct actually run. Very detailed models of a particular physical process. You gain more insight, you run an optimization. But computation is different from reality. So then you go build whatever you’re designing, do the experiments, take measurements, and iterate. All of these steps still have to be followed in the age of AI. Until recently, all of these steps were done in sequence by different people. A theoretician might take years on the first part. Then a specialist in computation runs the models and optimizations. Then a specialist in measurement handles the experiments. Then you put it all together, and maybe you have an outcome. But generally, you're waiting on each part of this sequence. You can't speed it up. Even before AI, some of that had been made more simultaneous. But it was still a long process. What AI has done is make all of it run at the same time. Especially with specialized agents. You can now do decentralized science. You don’t need to know everything about every domain. I still think expertise matters, but AI has accelerated the whole process."
English
2
1
5
8.2K
The Peel
The Peel@ThePeelPod·
I asked a top University of Michigan professor why there's been so much grade inflation at US universities: "I think there were many things going on. You could say that professors were catering to their customers, the students. The students would tell you they're working harder. You could agree that students are more prepared for college now than they were 20 yeas ago. I also think in most universities, the expectations is "if I work hard, I get an A". In many cases, effort is actually correlated with outcomes. If you actually work hard, you do learn. I don't want to be an idealist. But I think there's also an element of "I'm paying $60,000 for my degree". I'm not saying hence people get better grades. But it's naive to dismiss this factor entirely. I think it's many different things. Many students do work very hard. And it's not like every professor is just giving out A's for free. But I am seeing a movement where things are leveling off or even coming back down a bit. Harvard actually pulled this off. 3 years ago, 70% of grades given at Harvard were A's. They actually brought it down a bit. But now students are complaining, saying they're stressed and they're working so hard. So as with many things, I think all of these reasons are true."
English
1
2
5
14.8K
The Peel retweetledi
Chris Hladczuk
Chris Hladczuk@chrishlad·
Product velocity is everything - and engineers demo to the whole team on Friday night. if this sounds fun, apply @hanoverpark
English
1
3
18
6.1K
The Peel retweetledi
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
🚨This University of Michigan professor just gave his PhD students a Code Red about AI: "These AI models are now able to reason through things at my level of expertise, in areas that I'm one of the world's experts in. Things that would have taken me 4-5 months are now getting done in a weekend. Until December of 2025, I talked about the potential of these AI tools. But over the past few months, they've gotten as good as leading edge researchers. But there's still value in developing your own intuition and thinking. The best way to learn physics is by working through the problems. You develop a lot of intuition by doing things the slow way. If want to continue to be competitive going forward, you need to experience the friction of learning everything yourself while also mastering how to use AI."
Turner Novak 🍌🧢@TurnerNovak

New @ThePeelPod with University of Michigan professor Karthik Duraisamy Karthik co-leads U of M's newly created Institute of Agentic Computing. It's a central node for the OpenClaw platform and helps researchers and developers using AI to advance scientific discovery and engineering. This is Karthik's first public conversation going deep on the new institute. We talk about how AI has increased the pace of scientific research, two new discoveries announced yesterday at ClawCon in Ann Arbor, how universities actually work, how AI has impacted students and education, what's happening with college grade inflation, and the code red advice he gave students. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 0:25 The Institute for Agentic Computing 4:27 OpenClaw Foundation and Lobster Compute Company 8:19 How Universities actually work 12:33 ClawCon in Ann Arbor 15:24 Two scientific discoveries made with ScienceClaw 20:06 How AI is speeding up scientific discovery 25:42 Supporting AI and OpenClaw development 29:55 Why universities function like VC funds 34:29 How universities get money from the government 40:55 Why some academics believe AI is a fad 46:17 Biggest bottlenecks in AI today 49:26 How AI will change the world 53:10 Karthik's Code Red advice for students 59:19 Separating learning and doing 1:03:10 Ways COVID and AI impacted college students 1:14:53 How the role of universities is changing 1:23:21 Why college classes suffered from grade inflation 1:26:05 How AI is actually impacting the job market 1:32:49 Karthik’s advice for students 1:39:16 Winning two NCAA basketball national championships 1:43:04 Almost dying in the Grand Teton National Park

English
30
63
615
106K
erica wenger🏕️
erica wenger🏕️@erica_wenger·
Suggestions for the best AI auto-clipper for podcasts? Our team does so much of it by hand as of now. Can't get @descript to work well at all. @FlightCast requires porting over our whole distribution setup. Appreciate any tips!
English
15
0
22
3.3K
The Peel retweetledi
Turner Novak 🍌🧢
Turner Novak 🍌🧢@TurnerNovak·
TIIL a lot of academics think AI is just a fad "A lot of them tried it once two years ago. It didn't work very well in their specific domain, and they never came back to it."
Turner Novak 🍌🧢@TurnerNovak

New @ThePeelPod with University of Michigan professor Karthik Duraisamy Karthik co-leads U of M's newly created Institute of Agentic Computing. It's a central node for the OpenClaw platform and helps researchers and developers using AI to advance scientific discovery and engineering. This is Karthik's first public conversation going deep on the new institute. We talk about how AI has increased the pace of scientific research, two new discoveries announced yesterday at ClawCon in Ann Arbor, how universities actually work, how AI has impacted students and education, what's happening with college grade inflation, and the code red advice he gave students. Thank you to @Numeral, @FlexSuperApp, and @Amplitude_HQ for supporting this episode! Timestamps: 0:25 The Institute for Agentic Computing 4:27 OpenClaw Foundation and Lobster Compute Company 8:19 How Universities actually work 12:33 ClawCon in Ann Arbor 15:24 Two scientific discoveries made with ScienceClaw 20:06 How AI is speeding up scientific discovery 25:42 Supporting AI and OpenClaw development 29:55 Why universities function like VC funds 34:29 How universities get money from the government 40:55 Why some academics believe AI is a fad 46:17 Biggest bottlenecks in AI today 49:26 How AI will change the world 53:10 Karthik's Code Red advice for students 59:19 Separating learning and doing 1:03:10 Ways COVID and AI impacted college students 1:14:53 How the role of universities is changing 1:23:21 Why college classes suffered from grade inflation 1:26:05 How AI is actually impacting the job market 1:32:49 Karthik’s advice for students 1:39:16 Winning two NCAA basketball national championships 1:43:04 Almost dying in the Grand Teton National Park

English
4
3
11
6.2K