Hi, this is Kence Anderson again. I'm here with Mike Holden of OWSI machining and we're going to talk today about what things get made in the machine shop, about American manufacturing, we're going to talk about some of the challenges of running machine shop and how AI might be able to help. Thanks for talking to us today, Mike. You bet. Can you tell us a little bit about the history of your machine shop? We're here in a beautiful building from the '30s and it has a really rich history. It does. We started the business in 1931 as a Schlumberger R&D facility, develop products throughout oil and gas, and still today we still do business with Schlumberger and multiple other oil companies. What things do you make here? We make a little bit of anything. Basically, our capacity is steel products, mainly for the oil and gas world, but 30 inches in diameter, about 100 inches long down to eyeglass screws. No kidding. Yeah. So there's a huge variety of parts that are made, there's also a huge variety of machines that are in here. That's correct. So about 60 machines? About 60 Machines. Then how many people work here? We have 32. Got it. Tell us a little bit about the skills that it takes to become a machinist. I know there's a variety of folks that become machinist sometimes maybe straight out of high school, sometimes a little bit of college, sometimes it's some machining expertise. What does it take to get up to speed? It takes quite a bit of time, there's not as much as an apprenticeship in the high schools that they used to be, but basically as a guy comes in, they pretty much have to start themselves at the bottom, work their way up. But two to three years machining is just a minimum to get your feet wet, and then really it becomes a lifetime trade on their own. That's what we see. It's really consistent with other manufacturing that we see, it takes years, not months. Years. Then decades to perfect the craft. That is correct. It takes a long time. What are some of the factors that make it so difficult to navigate? I think one of the things I saw is there's a lot of different machines out there, but you also have a lot of different operations you're doing, right? They do, it's just as every day something is new to them, whether they're turning on a lathe, milling on a mill. Yeah. There's just in the tolerances today of solid modeling and stuff, what kids have done, it's just so much tighter than it once was. It really takes an awful lot of skill for what it used to be. You said what's splitting a human hair? We split a human hair 20 times. That's down to one micron. We can get there. That's one of the challenges I understand of running a machine shop is you don't get second chances. There's no second chances. There's a thing called scrap. That is correct. We do not get paid for scrap, you get paid only for good parts. In other words, when you come to us and buy your machine goods, you only pay for what is good, you don't pay for anything that's bad. Right, that makes a lot of sense. A lot of what these guys are learning over time, and I really like your philosophy for how you run your shop, in pods of operators that are running their own business inside the shop. We are trying to get there with AI where they can basically be their own boss and run their own self and get paid a commission on basically what they manufacture. What are the biggest challenges from a business P&L perspective in running a machine shop? Keeping the right work, and also eliminating any of your problems and being efficient. You don't want your time slipping away. Basically it is the time and materials type situation cost-plus, so that you use AI, you use what you can to make sure that every machine stays profitable at the hourly rate, that you're bidding at. It's a real challenge. How do you see AI or an autonomous AI helping improve this operation? Very easily. It can tell you where you've got your problems before you know it. In other words, it's calculating minutes, seconds every time on every machine. As soon as you get up out of your desk, you should know where you're going, where your problems are, also for maintenance, histories and things like that. It keeps up with all that. Yes. A lot better than any human mind can ever do. It's really interesting because walking around, I saw how you've instrumented up for measuring your whole machine shop, and it's a lot of people think of machine shops as not the highest tech, but you're using some pretty high-tech measurement in there. We do. You said the AI can be able to hear and perceive sounds. It can. Similar to the way humans can, but also you talked about the ability to tell you which operations are in the guard rails, which operations are outside of the guard rails, so you can make some really complex trade offs. That's right, what it should do for us what it does for us is that it calculates the time, it knows where we are per part, knows where we are at all times with the machines. Basically when you get up out of your desk, you're not walking around the 60 machines and wondering what's going on, you know exactly where your problems are before you get up and you hit them immediately and it just makes you far more efficient. Fantastic. Thanks for talking to us today, Mike. Thank you very much. I appreciate it.
