Looking For More?
Watch this episode on YouTube
Connect with the Manufacturing Industry here www.amtonline.org/events
Discover the past, present, and future of American manufacturing with www.mmsonline.com/madeintheusapodcast
Explore, watch, read, learn, join, and connect at www.imts.com/
Tune in to the AM Radio podcast www.additivemanufacturing.media/zc/am-rad…o-podcast
For the latest in Manufacturing Technology news www.amtonline.org/home
Ramia Lloyd:
Welcome to the Tech Trends Podcast, where we discuss the latest manufacturing, technology, research and news. Today's episode is sponsored by Modern Machine Shop's Made in the USA Podcast. Hi guys.
Benjamin Moses:
Hello.
Stephen LaMarca:
What's up?
Benjamin Moses:
How's everybody been doing? Let's get right into some articles. Alyssa, you got something from IBM, which is a blast from past. IBM hasn't been super relevant, but they got some stuff going on.
Stephen LaMarca:
My first computer was an IBM.
Elissa Davis:
Yeah, I mean...
Benjamin Moses:
Yours, or just in general?
Stephen LaMarca:
My dad brought it home from work.
Benjamin Moses:
He stole property?
Stephen LaMarca:
He stole property... I mean, they were getting rid of them. It had a CRT screen that was blue with white lettering. We all see in movies and media that it's like...
Elissa Davis:
It's like, green.
Stephen LaMarca:
... black with green or white text, like command prompt. This was blue with white.
Elissa Davis:
Interesting. I guess that makes sense, 'cause that's their colors, right?
Stephen LaMarca:
I guess, man.
Benjamin Moses:
Well that's good branding.
Elissa Davis:
Whenever I think of IBM I always think of Jeopardy where they went up against the IBM computer.
Benjamin Moses:
The Watson.
Elissa Davis:
Yeah, and they lost.
Ramia Lloyd:
And they lost.
Elissa Davis:
The did.
Ramia Lloyd:
[inaudible 00:01:00] that in there. Lets leave that in there like that.
Elissa Davis:
The now-host of Jeopardy, Ken Jennings, lost to Watson. I just want to put that out there.
Ramia Lloyd:
Wow.
Elissa Davis:
Okay. But yes, IBM, they've secured a patent for a 4D printed smart material, and it is... So according to the patent, these smart materials can use shape memory alloys or polymers that respond to external forces like temperature, light, magnetism or electrical currents. So after being deformed, the material returned to its original shape.
Benjamin Moses:
This is interesting, 'cause-
Stephen LaMarca:
Metamaterials and non-Newtonian materials...
Benjamin Moses:
Yeah. It's a term we've been thinking about for the past four or five years, of these materials that can react and then go back to the original shape. Smart materials or metamaterials. Metamaterials has kind of faded a little bit, I think.
Stephen LaMarca:
So it was a 4D printed part, a part that is never near-net-shape?
Benjamin Moses:
It's never correct. I actually that's a fair point. I think 4D might be the new term that we're probably going to use for these materials, going forward.
Elissa Davis:
So it says, "The design allows for delivery of microparticles between one and 100 microns in diameter."
Benjamin Moses:
Oh, interesting.
Elissa Davis:
"It can control methods, mean... "Its different control methods mean it can travel through various media, making it useful for medical and industrial applications."
Benjamin Moses:
Interesting.
Elissa Davis:
Yeah.
Benjamin Moses:
In our last committee meeting the material sciences came up as... particularly the computation required to experiment with materials, being able to predict on a screen or in a digital environment new materials. So this kind of extends where we're headed, and I think in terms of material sciences, where there's a lot going on for materials. I think we're going to see not quite Star Trek levels of innovations in the next couple of years, but a lot of innovation on either custom materials or how we apply materials.
Elissa Davis:
Like Big Hero 6 level.
Benjamin Moses:
Big Hero 6. Yeah, exactly.
Stephen LaMarca:
I want to highlight real quick what you said earlier, because I think that was really insightful, that I think 4D is the new term for non-Newtonian or metamaterial.
Benjamin Moses:
I think so.
Stephen LaMarca:
Probably because of branding issues with Meta, and non-Newtonian goes very anti-science. And this is science.
Benjamin Moses:
I like that.
Elissa Davis:
Yeah. So IBM, working to become relevant again, I guess.
Stephen LaMarca:
Very cool.
Elissa Davis:
Good on this.
Ramia Lloyd:
Bring Watson back on Jeopardy.
Benjamin Moses:
I want to brag for a little bit.
Stephen LaMarca:
I think [inaudible 00:03:20].
Benjamin Moses:
AMT was the early adopters of Watson. We developed the Tectrons app to help ingest research papers, and we used Watson to consolidate a full research paper into keywords, concepts and summaries. So we shoved all of our papers through Watson back then, so we'd understand financial models of doing these types of transactions...
So we took NSF, we took NIST, shoved in like 200,000 papers through Watson and we paid like, $1000.
Stephen LaMarca:
Dude, that's a great point. We should look back into IBM, see what they're up to.
Benjamin Moses:
No, no, no. No, no.
Stephen LaMarca:
Oh.
Ramia Lloyd:
They're up to this. They're patenting the...
Stephen LaMarca:
[inaudible 00:03:53] and say, "Thanks for sharing. This is amazing."
Elissa Davis:
Yeah. Yeah. So that's pretty cool. Yes, "The machine learning algorithm controls the printer material to transport microparticles practically anywhere."
Benjamin Moses:
One thing that IBM can't solve is daylight savings time.
Ramia Lloyd:
Oh my God.
Benjamin Moses:
Why are we still suffering through that?
Elissa Davis:
Oh my God.
Ramia Lloyd:
It makes me want to move to Arizona, 'cause they don't do it.
Benjamin Moses:
They don't do it.
Ramia Lloyd:
They don't do it.
Stephen LaMarca:
That's cool. I'm jealous.
Elissa Davis:
It happens twice a year, and still every time I'm like, "What's happening? Why does my entire nervous system feel completely out of whack?"
Ramia Lloyd:
Yeah.
Benjamin Moses:
It's so much work.
Elissa Davis:
It's draining.
Stephen LaMarca:
My wife and I had an existential crisis because we didn't know it was coming this past weekend. And while there's never a news announcement, there's never a notification... make your phone go bleep, bleep, like an Amber Alert or anything. "Hey, daylight savings time is coming."
Ramia Lloyd:
Amber alert.
Stephen LaMarca:
There should be one. But usually we hear from a friend. And it's not that we're like... I mean, we are a bit hermit-like, we are a bit recluses, but we do talk to people. And nobody knew. Nobody knew daylight savings time was coming this weekend.
Benjamin Moses:
The only reason I knew about it is 'cause I was listening to WTOP, the radio station. Which I do listen to radio, like over-the-air radio once in a while.
Ramia Lloyd:
That's crazy.
Elissa Davis:
I saw it on Facebook.
Benjamin Moses:
You saw it on Facebook?
Elissa Davis:
Yeah. And then I looked at my calendar on my phone and it said, "Daylight Savings Time." I was like, "Oh man." For me, I love the one in the fall where we get an extra hour of sleep.
Ramia Lloyd:
Heck, yeah.
Elissa Davis:
But this one in the spring, people are like, "Oh, it's great 'cause the trade-off is you get a longer... You get more sun during the day." And I'm like, "Yeah, but I like sleep." So...
Ramia Lloyd:
Yeah. For people who like to take midday naps, this is actually the worst time 'cause now it's super bright at my perfect nap time, which is like 5:30 or 6:00. And I'm like, "I can't, the sun is too much."
Benjamin Moses:
See... You were about to say something.
Stephen LaMarca:
I was going to say something about sleep. How come sleeping in is lazy, but going to bed early isn't? It's just a time shift.
Elissa Davis:
Well, because I think it's the difference of like...
Ramia Lloyd:
Ben's confused.
Elissa Davis:
I think people would argue that it's the difference between, and I'm not saying I would argue this 'cause I love sleeping in, but it's the difference between sleeping the day away and getting a head start on your day.
Benjamin Moses:
The best sleep I get is when I wake up... So my routine's really weird right now because I wake up, have a little cup of coffee, get Amelia up at 6:10. She continues sleeping. And then I have about 10 more minutes where I wake her up again at 6:20, and then she actually starts her day.
Elissa Davis:
Real.
Benjamin Moses:
That little 10 minute in between the first and second wake up with her is the best sleep I ever get. That re-sleep in the morning is... Those are good.
Elissa Davis:
That's real.
Stephen LaMarca:
I'm going to put my libertarian hat on real quick and just say, it sounds like daylight savings time is about control.
Ramia Lloyd:
Well, it is an outdated form. It has to do with farming, just like school days and work days...
Stephen LaMarca:
[inaudible 00:06:51].
Ramia Lloyd:
... all revolve around old school farming schedules.
Elissa Davis:
It's built on the back of child labor.
Benjamin Moses:
Speaking of outdated stuff, I want to talk about hot dog standards.
Stephen LaMarca:
I want to talk about hot dog standards.
So okay, you're going home from work, you're tired. May have been a good day, may have been a bad day. You want a quick bite, you want comfort food and you want minimal effort. Boiling a hot dog in beer, hot dogs in beer, and then throwing them on the grill. Anyway...
I like that as a hot dog consumer and purchaser at the grocery store, I like that the packages of buns are now relatively the same quantity as the package of dogs. It's nice, eight and eight, no difficult math to do, stuff like that. But it's 2025. We've got all this modern manufacturing and all of these standards in place. Why are the dogs still not the full length of the bun? You can get longer than the bun and you can get shorter than the bun...
Benjamin Moses:
But it never is correct.
Elissa Davis:
Yeah.
Ramia Lloyd:
'Cause you fill up on the bun.
Stephen LaMarca:
Where is ANSI and ISO on this?
Elissa Davis:
We should talk to-
Stephen LaMarca:
Brats the same way.
Elissa Davis:
We should talk to IBM, that's what they should be doing. Yeah.
Benjamin Moses:
Getting into the hot dog ISO standards?
Elissa Davis:
Yeah, getting in the hot dog game. Yeah, I mean...
Stephen LaMarca:
We all know that hot dog buns are better for brats too, even though the IDs and ODs don't match up. But I don't want a mouthful... Just a mouthful of bread and condiment?
Elissa Davis:
You don't like bread and ketchup?
Stephen LaMarca:
I need the prot.
Elissa Davis:
Just by itself?
Ramia Lloyd:
The brot. Please.
Elissa Davis:
Man.
Stephen LaMarca:
All right, I'm done.
Elissa Davis:
Man, I grew up... I ate so many hot dogs growing up, 'cause they were cheap. And like everyone knows, I have a very large family. And my mom said that apparently when she was pregnant with me she ate a lot of hot dogs, that's what she was craving the most. And then as a child, that's what she would make. I'd get home from school and she'd boil a hot dog and cut it up and that was my snack when I got home from school.
Ramia Lloyd:
Like, ketchup and mustard, that was my dad. My dad-
Elissa Davis:
Oh my God, yes.
Benjamin Moses:
Growing up we had the hot dog 10 different ways. We would boil it or we would chop it up and saute it. We'd put it on pizzas.
Ramia Lloyd:
Beanie weenies.
Elissa Davis:
Yes.
Stephen LaMarca:
Beanie weenies.
Benjamin Moses:
I know these words.
Stephen LaMarca:
Throw a can of baked beans in a pot, cut up some hot dogs and throw them in there. Beanie weenies.
Elissa Davis:
They were really good in mac and cheese, too.
Stephen LaMarca:
Solid in mac and cheese.
Ramia Lloyd:
No.
Elissa Davis:
Yeah. No?
Benjamin Moses:
[inaudible 00:09:12].
Ramia Lloyd:
I'm drawing the line right there, at putting a hot dog in my mac and cheese. Please.
Stephen LaMarca:
That's fair, because I don't actually think I've ever put it in the mac and cheese, but I've had it alongside.
Ramia Lloyd:
Okay, that's fair. That's a really solid barbecue staple.
Stephen LaMarca:
Makes it an American meal.
Ramia Lloyd:
Yeah. Grill your hot dogs...
Stephen LaMarca:
It's part of this complete breakfast.
Ramia Lloyd:
... make some Kraft mac and cheese. June through August, that's what hits. But don't mix them together.
Benjamin Moses:
Speaking of American meals, Ramia, can you tell us about today's sponsor?
Ramia Lloyd:
Hot Dogs.
Elissa Davis:
That was really good.
Ramia Lloyd:
We're sponsored by, what's the hot dog... Never mind.
Elissa Davis:
Hebrew National?
Ramia Lloyd:
I was going to say Portillo's.
Stephen LaMarca:
Saboret? Portillo's. How could we leave out Chicago?
Ramia Lloyd:
That's what I was thinking. Anyway, let me just read this thing.
Tune in for Modern Machine Shop's Made in the USA Podcast to explore manufacturing issues faced by companies making an intentional choice to manufacture in the US. Featuring commentary from OEM leaders, Made in the USA blends its newly century-long expertise with a unique audio storytelling experience to shine a spotlight on the past, present and future of American manufacturing. Find Made in the USA on Apple Podcasts, Spotify and all major podcast platforms. Follow Modern Machine Shop on Twitter, Facebook and LinkedIn.
Benjamin Moses:
One of the best parts about today's episode are my transitions.
Ramia Lloyd:
No, you're killing it. Absolutely killing it.
Benjamin Moses:
I really like my article that I found, because I want to tell a robot what to do for once. So I found an article, actually I think Steve found this, on the industry news posted there [inaudible 00:10:41]-
Stephen LaMarca:
There's going to be a solid write-up on it in the tech report.
Benjamin Moses:
From a Techsplorer, "A framework allows a person to correct robots actions using kind of feedback they would give another human."
So this is a joint project with MIT and NVIDIA, where you could correct a robot's behavior with similar interactions. So the concept would be you're on the floor with a robot, you're actually near the robot. So assuming collaborative robots. The method would be just giving it a little nudge. If someone's welding or if another human's doing something, most of the time you could visibly or verbally correct them or give them different directions. Sometimes you just have to point to the different thing or move it a little bit. So this is very similar, where they have a framework for correcting a robot and the robot inferring that information and doing that task based on that correction.
Some interesting stats. So they said they improved the success rate, was 21% higher than alternative methods not using human interaction. And they bring up a really interesting point, where some of this is kind of related to both the technology of the robot inferring this information and applying it, and the physical side of it of being able to take that information. But also the policies related to setting rules and applying these in different scenarios.
'Cause one thing they talked about is robot programming. You can give it a set of boundaries and the program will say, "This is a valid path," but just because it's valid doesn't mean it's the best path. And I think that's where the iteration that the article-
Stephen LaMarca:
Or free from collisions.
Benjamin Moses:
Or free from collisions, right. There's always going to be something where it's not part of the environment or the digital environment, right?
Stephen LaMarca:
Yeah.
Benjamin Moses:
So it's a very interesting thought process where they're like, "Don't retrain the AI model built to develop the path, train it in the scenario." Because it's too much labor. It's not feasible to retrain the AI model or machine learning model to develop that path.
Stephen LaMarca:
Absolutely. I think it's just so cool that they're talking about, you can change the code of your robot program just by... So you can whip up a quick base code and hopefully it's close enough to what you need, and then hit go at a slower speed. And as it's going I guess you can engage like a teach mode or an active teach mode, so it's doing what it's doing or what it thinks it's supposed to be doing, and you gently correct it along the way. That is so cool. I love that.
And I think what this article really highlights the most is how this is the next step in transitioning from programming machines and robots from a line-by-line or just hard-coding programming method to being directly involved, in some cases physically involved, with programming your device. It's so cool. I love this. I don't want to use computers.
Benjamin Moses:
It reinforces the current trend of making the human-to-machine interfaces more conversational, more applicable to how humans work. 'Cause if you go back a bunch of years ago, learning a language to do something is a not the best way to implement manufacturing technologies. So this gets us more to rapid innovations on the floor. So...
Stephen LaMarca:
I think it would also be super cool if you could E-stop a robot, just be like, [inaudible 00:14:07].
Elissa Davis:
I'm just thinking of, I don't know, when it's talking about training I picture a Major Payne type character yelling at the robot. I know that's not what it's like.
Benjamin Moses:
Like a drill instructor.
Stephen LaMarca:
"He ain't dead, but he ain't happy."
Ramia Lloyd:
I kind of imagined something similar, but for some reason my mind went to baseball, where the coach is just like, "Get it, get it, get it. And then he smacks the robot on the butt.
Elissa Davis:
And the robot-
Ramia Lloyd:
Like, "Good job."
Elissa Davis:
"There's no crying in manufacturing."
Benjamin Moses:
Yeah. That's such a good [inaudible 00:14:36].
Ramia Lloyd:
"Get on base."
Stephen LaMarca:
And the robot turns towards you, "Excuse me, do not touch my joint on. That is my joint one, and it's a no-no zone."
Benjamin Moses:
Steve, tell me about Sony's AI. I know you've been doing that experiment with Gran Turismo, I want to learn more about it.
Stephen LaMarca:
Yeah. So now that I'm playing Gran Turismo all the time and I can't shut up about it, I did notice that what I thought was this gimmick, that Gran Turismo was advertising Sony's Sophy AI in the game. And the reward, the payout, the credits that you receive in game aren't very high, but I like dabbling with stuff. Once I've completed all of the races that I want to complete, it was like, "Let's experiment with... Let's see what this Sophy AI is all about."
And at first I didn't really notice much, but everybody knows how bad the general computer opponents are in Gran Turismo. Like, there's some... It grades on a curve. If you put down a really solid lap time and are far ahead of the AI, it tries to keep up with you. And in some cases, as you approach the end of the race, all of a sudden the computer players behind you start putting down killer lap times, and they just take a steaming dump on your fastest lap and beat it by two and a half seconds. And it's like, well, where was this at the beginning of the race? If you are doing this now, I shouldn't even be touching first position.
They have since changed it, because it used to be on an algorithm where they tell the computer opponent what the best driving line is around the track. And this is the car you have, they program the computer to the car and just go. And we slow down the lap time or speed it up depending on the difficulty that you choose. And there's some variability to how good of a racer the player is, that it adjusts it on the fly.
The AI is entirely new. They don't have any programming on the AI in the game. All they give the AI is the track boundaries, where's off track, where's on track? Like the left side, the right side. And where the other cars are, and the performance of the car that is given to the Sophy AI. And they've gone through several iterations, and they've been some hits and misses here. It got to a point where...
So the AI was learning the track on its own and just putting down insane lap times that nobody could touch, and it was regularly happening. There's a lot of players online who can, if they manage to start ahead of the AI, they can keep the AI behind them because the AI's primary focus is sportsmanship, not taking out other cars. And if you can stay in front of the AI or if you start in front of the AI you can finagle your way to stay in front of it. But if it gets ahead of you, the AI out-brakes you every time. It always knows the optimal braking zone, it has complete control of the brakes. So where you're leaning on ABS or you're constantly locking up the brakes and fluctuating the brake pedal on your own, it knows the perfect limit of the brakes in any situation. You cannot out-brake the AI.
I want AI-enabled brakes, by the way. And then... I think I'm tapped out on it. Yeah, it's just so cool.
Benjamin Moses:
It's weird you mention this. I would say about seven years ago, DARPA actually did a test where they were testing humanoid robots to basically walk a path. Very similar to your scenario, where it's like, these are boundaries, go do a task. But it was more of instructional tasks that were given to it and had to figure it out. So it had to figure out how to walk but also stay within the boundaries.
So this is one of the previous iterations, before Atlas and all that Boston Dynamic robots. But it was very similar where the early iterations, the robot would topple over or they would get stuck in a loop and they wouldn't know what to do.
Ramia Lloyd:
Same.
Benjamin Moses:
But it would learn pretty quickly, that... One or two iterations in the competition later on, where it was able to actually execute a lot of these things. I feel like this is a very similar application, where in the digital world it's able to iterate very quickly to get to something fairly useful.
Stephen LaMarca:
Yeah. The really cool thing about the Sophy AI is also, because they put such an emphasis on the AI sportsmanship... Oh, I remember the other thing I wanted to say now. But because they focus on the sportsmanship so much, it's really optimized how good the AI is at navigating slower traffic.
Benjamin Moses:
Oh, yeah.
Stephen LaMarca:
Like, the way it dips and dodges and weaves in and out of traffic to get around slower cars is brilliant. It's so well done.
But it also made me remember that one of the iterations of the Sophy AI was so good, the AI learned how to handle the cars so well and was getting so fast with literally inhuman speeds, lap times, that you would watch the replay, the race replay, and focus on the AI to see what it's doing. All four wheels are smoking the entire time. They have since changed that by implementing the tire wear on the AI to keep it from nuking its tires within the first couple of laps. But also because of that, they also realized that... They used the AI to find broken areas in their physics algorithms.
Benjamin Moses:
Oh, that's cool.
Stephen LaMarca:
So there were some issues with the suspension tuning that they found through the AI being able to put down the optimal lap times.
Benjamin Moses:
I think the biggest thing I really like are the boundaries that they set up. Like this concept of the sportsmanship in their agent is really interesting. I mean, that's come up quite a bit in some of our committees, about being able to enter certain markets and the boundaries those companies set in terms of where they're allowed to go. I think putting those boundaries on AI agents in terms of, "Don't do this thing," or allow it to do more on that...
Stephen LaMarca:
That's a good point. I overall just like the sportsmanship concept that Gran Turismo has, because if you are a trash player, not bad but bad to other people, if you're dive-bombing corners and purposely missing your braking zones, just to rely on crashing into other cars to let them help you around a corner, they're going to give you a really bad sportsmanship score and they're going to make you play with other trash players. But if you're really good and you never contact other players and you're happy sacrificing a few positions, you're going to play with other kind, non-trash people. It's really cool how they separate... It's an interesting... It's a classist system that I totally support.
Benjamin Moses:
Awesome.
Ramia Lloyd:
I don't think I understand how you play Gran Turismo.
Elissa Davis:
I don't... Yeah.
Ramia Lloyd:
This entire conversation I was like, "Yes." And then I was like, "No."
Benjamin Moses:
Very fair.
Stephen LaMarca:
The idea is you're supposed to play it like it's your actual car. Would you crash into other people or purposely crash into a speedway wall to go around a corner? No, you're going to have to pay for that.
Elissa Davis:
I think I got lost when you were talking about traffic? 'Cause it's...
Stephen LaMarca:
So if there's slower cars on the track or if there's multiple classes of cars, like a GT class versus a prototype class on the same race, the prototypes have to weave around the slower cars. Because eventually they get lapped. And that is an aspect of actual endurance racing that's really fun.
Ramia Lloyd:
But you're not playing on a controller.
Stephen LaMarca:
I play on a controller because I'm poor.
Benjamin Moses:
Couldn't buy a steering wheel?
Ramia Lloyd:
But you can have one of those like-
Stephen LaMarca:
I don't have one of those rigs.
Ramia Lloyd:
Oh, okay.
Elissa Davis:
Oh, man.
Ramia Lloyd:
I was confused when you were like, braking, and I was like, are you sitting... You know, like you play in the... Not the amusement parks, but the game rooms or whatever.
Elissa Davis:
Yeah. Arcades?
Benjamin Moses:
Oh, arcades [inaudible 00:22:54]. Yeah.
Ramia Lloyd:
I'm just imagining you in an arcade booth, just braking and driving.
Stephen LaMarca:
Next time. We'll work on that. Yeah.
Ramia Lloyd:
Yeah.
Elissa Davis:
I'm picturing that movie Ford v Ferrari. Did you see that movie? Yeah.
Stephen LaMarca:
I love that movie.
Elissa Davis:
That's what I'm thinking.
Ramia Lloyd:
And I've also seen the Gran Turismo movie, so I'm expecting you to be like that guy who's like, "Vrrrrr."
Benjamin Moses:
Yes, that is [inaudible 00:23:11].
Ramia Lloyd:
Yeah, okay.
Stephen LaMarca:
That's me.
Ramia Lloyd:
I love that.
Benjamin Moses:
Ramia, where can they find more info about us?
Ramia Lloyd:
techtrends.amtonline.org.
Benjamin Moses:
Nice.
Ramia Lloyd:
I'm glad we got that right, now.
Stephen LaMarca:
I didn't even know it. Good job.
Elissa Davis:
[inaudible 00:23:23].
Stephen LaMarca:
Oh, everybody. I was in those meetings.
Benjamin Moses:
And don't forget, [inaudible 00:23:29].
Stephen LaMarca:
That's right. Oh, we should have said it earlier. Bing-bong.
Elissa Davis:
Bing-bong.
Ramia Lloyd:
Bing-bong.
Looking For More?
Watch this episode on YouTube
Connect with the Manufacturing Industry here www.amtonline.org/events
Discover the past, present, and future of American manufacturing with www.mmsonline.com/madeintheusapodcast
Explore, watch, read, learn, join, and connect at www.imts.com/
Tune in to the AM Radio podcast www.additivemanufacturing.media/zc/am-rad…o-podcast
For the latest in Manufacturing Technology news www.amtonline.org/home
