BehindTheTech: Interview with VP at Leardon Solutions
The following is a transcript of an interview between Murray Learmonth and Jonathan Emig. Murray Learmonth is a former HP employee and the owner of a successful product development company.
Video of the Card Shuffler Prototype being Demoed
Murray Learmonth is the owner of a successful product development company, Leardon Solutions. The following transcript is an interview between Murray and I discussing his business and some of the products he's built. The interview took place in his office located in San Diego, California. His most recent project is a card shuffler built for a gaming company. They signed up for a deadline to build multiple prototypes for a trade show in Vegas.
What was the hardest part of this project?
Well, I think the challenge was really the schedule. We had a schedule that was unforgiving because we had a trade show that we had to be at in the middle of October and there was really no way around that. And, we had to sign up for that schedule as well.
Did it all work out?
Yeah, it was down to the wire.
How many did they ask for?
For the show they asked for 8 units.
How long did it take?
It took about 1.5 years, but that's normal for a new mechanism at HP it would take about 2 years.
And so how do you delegate tasks?
We have subsystem responsibility. So this is called the input subsystem. Electromagnetic, drive mechanism. So one engineer was working on this. One engineer was working on these two systems.
Because I had done a lot of paperwork before, my responsibility was the main mech.
The core mech so I had the core mech responsibility
We had one electrical engineer who was responsible for all the electronics.
It goes through the deck and looks at every card and makes sure there aren't any doubles.
You break it down. You work back from: this is what we need to deliver. We had to be ready for the show for the middle of October. We had the first built units around the middle of September.
How many iterations did you build?
We made 3 iterations of prototypes. One generation was metal which got certified by the gaming commission for randomness and the third generation was SLA.
And, that's never been done before in a high speed paper moving device. At HP you used to have to machine these pieces from plastic, and then test them but we didn't have time. To machine one of these takes like 2 or three weeks. We were spinning this in one day. So we were getting turn around on these in 24 hours. So we had no time in the schedule so we had no option but to go straight to this and build-test-fix build-test-fix.
So to clarify, that piece was 3D printed, and that's not something that's typically done?
So the main reason we don't typically 3D print is because the stability isn't there, but now you're getting close to resins. I mean this was a new resin the company that we worked with. And this resin was pretty stable for what we needed. So it worked pretty well. Not perfect, but it got us much closer to what we needed. There just wasn't any forgiveness in the schedule.
But, based on the performance and how it went at the show we're actually going to re-design this for production. We're actually going to launch production tools in the middle of next month. We're going to be building production representative units. Which we call them LP (lab prototypes) but they're production representative units.
And, they're the first tool parts that we'll be building, 25 of them next march. Then by June or late may we'll be building 50. Which is really for the factory to get their processes ready.
So even from the prototype stage which is October to being able to sell units is August.
Were you guys full time on this project?
We had full time people on this project most of the time, not all the time because they'll peel off. Because, once the design is done then you're waiting on parts so people are not sitting around waiting on parts.
But, you know, in terms of manpower you need at least 2 mechanical engineers, 1 electrical engineer, and we had sort of 3 firmware engineers as well. We had 2 on the main mech and 1 on the vision system.
And then an industrial designer as well for a short period of time at the beginning.
I think at the peak we had 10 people working on it.
Is this a prototype?
This is actually before alpha, this is called breadboard or BB stage. There's many different cycles we'll be going through.
We look at these as test beds. We did test bed 1 test bed 2. And then we'll do a release for tools and then Alpha and then once we get tools it'll be Beta.
Is everyone in the office?
Most of the people that worked on this one were actually US based, so at the time of the last 4 to 6 weeks everyone was here. Because everyone was building up units and contributing.
So is it slower right now?
No, it's slower in terms of what we gotta do here in terms of testing, but everyone is working on the redesign of this, the firmware redesign, and the integration of the vision system, so but for that everyone can work in their own silo. They don't need to be here every day. The firmware guy can sit in his pajamas at home. The vision guy is working on his stuff. So that's pretty much the way it's going.
How do you handle the tooling process?
So when we release the design for tooling, what we're gonna do is, generally when you release a design for tooling you dont say, well I hope it's gonna be okay and the tools are gonna be ready in 2 months and the parts are gonna come off and it'll be great. Once you release the design for tooling you usually prototype it. You either machine it or you actually get SLAs and build up that complete design one more time just to make sure it functions. And if you realize something is wrong you change it quickly just so you don't wait till February when the tools are ready and go aww we gotta change the tools now. It's kinda like you're last look before production design to verify it and make sure it's gonna be okay. And, there's always stuff that's gonna fall through the net, but it gives you your best chance to be perfect.
Do you want to expand?
I don't think we want to go crazy. As we've said in total we've got 16 people in payroll between consultants and part time consultants and...
It's really about the skill expertise levels. There's a couple of people I want to hire. And it's gonna be based on their availability. A couple of people are actually gonna leave the companies their with. And they have expressed an interest in coming to work with us because they like the flexibility of not having to go to the office everyday. They're working on diverse projects.
They're both in corporate jobs right now and they don't want to go back to sitting on a desk, they want a different lifestyle.
So yeah, I think there will be organic expansion, but I'm not going to force it. We're probably going to expand maybe 1 or 2 more people at a time in the next year.
Have you always had an interested in being entrepreneurial?
I had been freelance for many many years before I worked for HP. I was a freelance consultant for IBM, the auto industry in Germany, and actually Boeing as well. And I had done that for well, gosh, maybe 10, 12 years. And so I had done that and decided to take a job at HP.
And you're a consultant as well, you get to be in control of your own destiny, of getting work, or networking to get the work. So this was like a logical extension of it. And so, entrepreneurial, I don't know. Are consultants entrepreneurial? But, they kinda have to be though because they have to generate their own work based on their reputation and their competence.
Did you ever want to try and invent something and patent it?
You can force an invention, that's the way I look at it. We have the team that we could execute and build that invention, but I've never thought, choose these for an invention to do. If you stumble across something then great, but we never really thought oh we're gonna do it.
But, I have talked to some people that are invention forcers. I met this guy once, at one of these networking events and he was one of those forced inventors. He basically looked at a market of what do people do, what do young people do, and it was gaming, and he said alright, I'm not gonna develop a game. What are things that people do during gaming that a high percentage of them have to do? So he kinda looked at this and thought they gotta eat. So he developed a food called GamerFood, which was chips that don't get food on your hands. So this guy had no interest whatsoever in anything other than forcing an invention or being a serial entrepreneur. He wanted to make money and he said, okay how am I gonna do this. So he worked with food scientists to develop flavorful chips that doesn't get crap on your hands so you don't get it on your xbox controller.
What's the biggest challenge you've run into?
Getting customers to pay. Technically it's never a big issue, but getting customers to pay has been a problem. We need to be more clinical in terms of invoicing.
Another problem has been working with liquids. Liquids and gases. You just can't control the behaviors of the gases.
Other than that there haven't really been any big challenges.
There have been a couple of small projects that we have had some problems with. For example we make them a prototype and maybe it doesn't do 100% of what they wanted, but we explain to them that yaknow James Dyson build 20000 prototypes to get there and you want to get it to the park in one?
And we do a pretty decent job of explaining to people that you want to do all this, but it takes time to do all this. You're not gonna get it right in one time.
But other than that nothing really has been a big issue. There have been small blips, but nothing big.
What is the biggest success?
Jeez, I don't know... probably getting this done is such a short space of time from a change of scope and design was probably the biggest one. But, out of all the other stuff, we've developed refrigeration control units and manufactured 50000 of them. And, AVACEN, the heat warming thing, we've manufactured about 10,000 of them. You know, so these are all pretty successful products.
The stuff we've done for big medical customers we've designed and build diagnostic machines for them and prototyped them.
The BrainOS, the autonomous vehicle is pretty successful. For this project we're working on a retrofit. They want to have cameras and a sensor system. So when it goes down the aisles it's scanning the store. And when this guy is autonomously going down the store it's cleaning the aisles and taking picture of shelves to know what it needs to stock. And it can send that out. And there is another autonomous truck that comes out and drops off the stuff.
Have you had any unhappy clients?
Oh yeah, when you work with lots of clients there's a certain small percentage of them that are pretty bad. That's why I like working with a large company. When you work with a larger corporation they've got standards of conduct that they're all adhering to. Individual are passionate. They've got a certain widget that they've been working on for the last three years it's their baby, you can't criticize you can't make changes to the design.