Bob: My time in the pharma industry, I've been in it for about 35, 36 years. I remember my early days, you know, there was barely a computer and we had no networking. I had a small little laptop that I just connected and typed some stuff and I could email and that was about it. We've come a long way and that's the beauty of technology, right?

Announcer: You're listening to Augmented Ops, where manufacturing meets innovation. We highlight the transformative ideas and technologies shaping the front lines of operations, helping you stay ahead of the curve in the rapidly evolving world of industrial tech. Here's your host, Natan Linder, CEO and co founder of Tulip, the frontline operations platform.

Natan: Welcome back to Augmented Ops. Today, I'm here with Bob Buhlmann. He's AstraZeneca's head of quality digital and computer strategy. He's a truly progressive leader in the digital space for transformation of pharmaceutical manufacturing. Hi, Bob. 

Bob: Hey, Natan. How are you today? 

Natan: I'm good. Good to see you again.

And, Bob, what is your job, actually? Explain to us what you do. 

Bob: Sure, Natan. It's been quite a journey. My time in the pharma industry, I've been in it for about 35, 36 years in quality. And over the last four years at AstraZeneca, I've been doing a lot of research. What I do is I work for what we call our central, think of it as like a global quality organization, central quality.

And what we do is we look at bringing in new technology, digital solutions, how we validate them, ensure the integrity of our data with those, and then all of the great things we can do with data and analytics. It's, uh, quite a job, but it's, uh, it's fun. There's a lot of cool things we're doing, a lot of, a lot of good new technology using stuff like Tulip, you know, how we're designing that and implementing that at AstraZeneca.

So it'd be great to discuss some of those things. 

Natan: Yeah, no, I'm looking forward. And we had a few episodes on quality and I think, you know, of course we're talking about it from the perspective of pharmaceutical, goes without a say. The topic quality actually transcends all manufacturing and, uh, at some point early in Tulip's life, we decided to get ourself a QMS and go back on the ISO 9x path and do all those things so we can actually operate in regulated industries.

And that, that was like a real big decision for the company that we had to balance with, uh, what startups love to do, you know, like Move fast and develop and agile and all those kind of things that, they're kind of contradicting like quality, especially if you think about, you know, you, you said 30 something years, um, it's a different world.

Bob: Very different world. It's a different world from last year. I mean, things change and technology moves so fast as you know. And so we did back in the nineties, nowhere near what we do now. So amazing. 

Natan: If you go back and you think about like, the 90s for a second this may be kind of a weird question from a quality perspective but what didn't change since the 90s despite all this technology and you know all those things what stayed the same what's the quality system constants that we have to keep in mind 

Bob: Yeah, technology changes, but the processes have all been the same, right?

There's still change control, there's still deviation management, product complaints. So our quality system, for the most part, stays the same. A little more harmonized across the globe now, right? So the Europeans, the Americans, right? That they're trying to align on things through things like ICH and stuff.

So a lot of that stays the same. But technology is what really changes it, right? And I remember my early days, you know, there was barely a computer. We had no networking, you know, I had a small little laptop that I just connected and type some stuff and I could email and that was about it. 

Natan: And that was the, the tools to manage a quality management system that approved a drug for 

Bob: market.

Yeah, it was amazing. So, we've come a long way, and that's the beauty of technology, right? Yeah, 

Natan: the first principles of quality have not changed. And the tools, things like deviation control, things like that, those have not changed. It's just technology changed. People are still people, I think, same. Same, yeah, same.

Bob: Yeah. I mean, you go back to your quality systems, like Plan, Do, Act, Check, right? Of, uh, Deming models and stuff like that. Those have been around forever. It's the same. The same. And the people are the same, like you're saying. But technology helps us look at it differently, right? It helps us uncover things, things that we may not have seen in the past when we're sampling product that maybe the technology has enabled us to pick up things that we weren't picking up.

Yeah. You know, faster releases, because ultimately we're trying to get product to our patients as fast as possible, right? 

Natan: Without risking quality, you know, safety. And it's just like such a juxtaposition between businesses want time to market and all that kind of stuff. But in some businesses, it's like time to safe product to market.

But this is really brings up the question. It's like, we know technology could be. You know, game changers, speed stuff up, uncover tons of value and all that kind of stuff. But the reality is like, pharma companies were slower to adopt technology as it comes to generally transformation production process.

And I think it's fair to say. We'd love to all our quality people, also quality department were sometimes taking their time. How do you explain that? You guys don't like tech? 

Bob: No, we love tech. We love tech. You know, a lot of regulatory agencies even say the same thing, right? Well, they're not adverse or opposed to new technology.

And we always hear, you know, here's a new AI tool. You know, the FDA is already using this. 

Natan: Yeah. And 

Bob: it's like, yeah, but they're not releasing products, right? So it's a little bit different. So there are, there are things, right? There's regulatory challenges, right? I mean, It's a very highly regulated industry.

You bringing a new technology that maybe they're not even familiar with it. They don't even understand. They don't know how to inspect it. Right. So that that's a hurdle, 

Natan: right? 

Bob: I mean, there's financial barriers, right? For sure. Sometimes you guys put out this technology and the cost is just way too high for folks, right?

We're not like the food store. You know, you go get a gallon of milk and it's 2 one day. And then somebody says, now it's 9 the next day. Well, we can't do that with pharmaceuticals, right? We can't just change the price of drugs. So there's financial barriers, right? Things of that nature. And a lot of it's the risk aversion, right?

Are there concerns about disrupting what we're already doing? Right. And sometimes when we change things, we have to refile the product, right? So we file for product approvals, a certain way of manufacturing the drug. So if we're going to take new technology to do something, it may in turn Say we have to go refile and that could take time as well.

So there's many, many things that get into it. What I like to say is that internally, you know, a lot of times the roadblocks, let's use quality since you want to pick on quality at the time. That's okay. Right. We don't want to change. We like doing the way we're doing it. You know, I come to work. I know how to do something.

I don't want to change. I'm uncomfortable. And that's really the issue that we have to deal with is how to get out of our comfort zone to make the news on our comfort zone. So that's 

Natan: not just a technological adoption barriers. It's in fact a cultural issue as well. 

Bob: It is. And when you get into cultural transformations of things, that can get into how do we adopt the technology, right?

And you have competing problems with like an IT organization saying, well, we're going to take everything. We're going to move it to the cloud. Well, okay, you know, what does that mean? And you know, we can't touch it anymore. And so everybody gets nervous because they don't really understand what's going on.

Or how can I go from my nice quality system, management system that I have at my site. And then I want to make it more standardized and global across the enterprise. So where each site has their own quality system using their own tools. Now, somebody like my group comes in and says, well, let's go do it across all of AstraZeneca.

And people get a little nervous with that because they can't touch it anymore. Maybe they're not in control of it anymore. And that's. Becomes an issue. One of the biggest issues though, really is, is if you have leadership commitment or not, right. The more, uh, up the food chain, we have commitment to it and they support it.

Then the, I think the easier it becomes, but there's so many benefits to it all by doing it and changing your culture. So. 

Natan: So, more recently, it's like, can we adopt more Agile, Scrum like methodologies, because I found quality and from URSs to Kappas and everything in between, kind of very waterfall y. Lots of documentation, SOPs, which is really coming from the ISO, uh, stem, you know, and the derivative.

And so, you come with, like, Agile cloud and all this kind of stuff. Even before we get to AI and all sorts of things like that. And I was trying to think about this and I was like, what was the last big wave that was like, uh, this kind of, um, friction? Was it like when the PC suddenly emerged? And he wanted to put that in a quality office and people were like, no, no, no, no, no, you don't understand.

We are totally fine with our whatever mainframes and binders and we don't need this. This machine is untested. I'm just trying to imagine like, what did people say back then? 

Bob: Yeah, it's funny. Again, it goes back to, um, we were working at a, at a plant startup back in the nineties and a lot of it was just manual.

Very basic, you know, relays. You press a button and the oven starts. Oh, look at that. Okay. It's, I can do that at home. It's just pressing a button, you know, but they, they use the, well, it's a, it's a, an integrated PLC and it's gonna give you an alarm. Okay, great. You know, it's just relay logic and it's all right.

So this is really cool. I was consulting at the time and I go to another, uh, job to help out on something and I go there and they're pressing tablets and it's a manual tablet press. The person's cranking. 

Natan: Yeah. 

Bob: And they have to crank so fast to put out the tablets manually. I'm like, wow, that's crazy. And as I go back to the other plant startup, I was telling people about that and they said, you know, guess what?

We just got another job to start another company, uh, plant and they're going to be fully automated. And I'm like, wow, that's going to be cool. And we go there and what they have is they have a formulation tank that can open a valve and send a product over to the next tank. 

Natan: Wow. 

Bob: Like, whoa, that's really fully automated.

This is really cool. And you can see it on the computer. There's a visual of it. 

Natan: Uh huh. 

Bob: It's like, wow, we're really growing now. So we're going from crank tablet presses, right, to starting and hitting a button to see an alarm. And now we're going to visually see things and they're calling them SCADA systems and stuff and things are really starting to flow to now.

I mean, it's just, it's unbelievable the technology now and what's out there. You want everything automated, right? Manufacturing of the future. Autonomous manufacturing. 

Natan: Yeah. 

Bob: So, really cool stuff. 

Natan: So, knowing all you know now, and seeing those waves in the adoption, if you could have a sit down with the head of the FDA.

And you would be like, uh, listen, here's what we need, the practitioners, to adopt digital technology better, faster, smoother, and still in a safe way. What would you ask the head of the FDA? 

Bob: Yeah, that's an interesting question, right? You know, their job is coming in on behalf of the public safety, right? So they have their backs against the wall, otherwise everybody will be suing the government, trying to sue the government and the FDA for not doing their job.

And they try. Maybe they're understaffed and they need more help. Um, and I think they do, they, they've tried and they, I know they do it at companies in AstraZeneca, right? To come in and learn, you know, learn with us, let's partner more. And we try to do a lot of that at AstraZeneca, partner with agencies on what we're doing, the new technology, continue to be open to that, you know, and sometimes I think they have everybody on the same level, right?

And it's like, here's a startup companies, the same as AstraZeneca has been doing this stuff for years. So, you know, give us a little bit of a break. Is there a scoring model or something we can do to put in, how can we help show that, you know, we are in control? And I know they've talked about this in different forums and stuff with FDA about, can we grade people based on data and look at things so we can focus our inspections?

I don't have any one big thing I would ask them other than just continue to partner with us, you know, and, and help look at a lot of this new technology together. And let's not be fearful of it. Yeah. And they have a tough job. 

Natan: You know what I would ask? 

Bob: What's that? 

Natan: I would ask, let's make a task force to adopt techniques from the open source community.

To implement in how we think through validation. What I mean by that, I'll give you some examples. Like when you think about the kernel of the Linux operating system, why is it so good? Because the open source process, even though it's not controlled and regulated in the same fashion, a regulator does that, but is in fact regulated by the community.

And, uh, this community, because there are experts there, and there's a process that is tested in test suites and automation and many, many other ways before stuff gets released. And even when it gets released in the mainline, like when people want to actually use it, they take the responsibility. So there is continuous quality validation that happens after that community process that ensures that the kernel goes into a certain level of quality.

And then, of course, if there's a deviation or a bug, you know, The whole community gets to fix this, especially, you know, the most famous stuff. Oh, we found a security issue in blah, blah, blah. Here's a patch. You got to do it. So, you know, when you think about this example, this is at scale quality that works and it involves a lot of open sharing over platforms and communicating in a way that is, I think, non traditional, not natural for pharma, but is a source of inspiration.

So that example one. The other example I have is like. In cybersecurity, there's all these repositories of, that collect all our threats all the time. And all the antivirus and endpoint protection and the cloud protection. Everybody's like tuning into those databases, and that's how you know what to defend against.

And you take that away, and then, to me that's like a form of information sharing that improves the validation of the software that is supposed to protect us, your firewall, what have you. You know, but I don't think the pharma industry is doing enough of that. 

Bob: Yeah, I mean, they're interesting points, right?

Ultimately it gets down to who's responsible. 

Natan: Yeah, 

Bob: right. So the drug manufacturers, right? We're responsible for the manufacturing and release of their product and if they can't control some of it, right? They can't control the open AI How do you govern that? How do we make changes? You know, we could be introducing problems and we're not aware of them.

And, 

Natan: you know, 

Bob: it gets into a lot of that, but there, I mean, there's, there's, those are ideas, right? Great, great. Those are ideas we can look at and continue to, to use it. You get into, um, you know, using open AI, it's, it's again, it almost becomes a cybersecurity issue as well as how are we using it and opening it up and allowing it to come in and stuff.

Natan: Yeah, lots to do to kind of contain those LLM based systems, but I will really want to talk about the here and now for a second, because I think, you know, the work we've been doing together, we've been trying to change the traditional validation, you know, speed it up and kind of introduce concepts of composability, And citizen development, democratizing the ability to build digital solution.

And that naturally creates a validation challenge. What was your approach to that when you were tackling this at Astra? 

Bob: So validation is an interesting thing, right? It's all about intended use, right? So you can buy a piece of software and say, here, I'm going to install the software. Yeah, but did you configure it and set it up right for you to use?

And that becomes the, the challenge. And then everybody wants to come and say, you know, we have to follow this process, like you're talking about before, waterfall process, even an agile process. You can use whatever you want. Just follow the process, just follow the process, right? You should be defining how you intend to use the system, and then we should be verifying it through testing and verification that it actually works.

And what we. What we tend to do in the industry is we tend to all go one way because some group or organization has told us to go that way, and we're waiting for somebody to say it's okay not to do it this way. Right? And we write all these documents. What's a 

Natan: concrete example of that? I mean, we love all the organizations, so you don't have to put anyone under the bus and who's telling us where to go left or right, but what's an example of like, here's something you should do, this is how it's done, until they say, oh, by the way, here's another way you can do it or stop doing that.

That would be a concrete example. 

Bob: Well, so you see things like, let's, let's go with the basic approach, this waterfall approach that was out there for years, right? You're going to write a validation plan, write requirements, write a spec, do a functional risk assessment, do a system risk assessment, do an IQ, an OQ, a PQ, a final report and goes on and on and on.

And then you get the Agile one. You make 

Natan: it sound like it's not fun. I don't 

Bob: know. It's not fun. It's not fun. I mean, it's, it's, it's, it's boring. It goes on forever and everybody's waiting for validation to finish so we can start to make product. Right? Then you get, well, we're going to go Agile and Agile is all the same stuff.

You just try to do it faster. It's like, well, what's changed? What's changed? Nothing's changed. Oh, we're going to write user stories now. Well, what's a user story? What's your requirement? Well, for me, the user story was written like a two year old. It's like, if you were an analyst, you should be able to do this.

Okay. The way I used to write it was just do this. 

Natan: Yeah. 

Bob: So what's the difference? And so Agile is just speed. We can do speed. I still want to see how are we defining what we needed to do? And there's a lot of concepts and things, and then we're waiting. And then somebody comes out and says, well, you know what we're going to do?

And the FDA is behind this. We're going to have computer software assurance, and that's a risk based approach, and we'll be able to do it. Okay, so what do you want to do? What should we write? And everybody comes back and you write the same stuff. So we didn't change anything. 

Natan: So it's kind of like a vicious cycle.

How do you break that? I think hot topics are like at risk validation. I guess like when it comes to platform, you know, validate the platform so you can have multiple use cases on top of something you already built. I'm kind of, I'm using it carefully because it's not like pre validated, it's actually fully validated.

Maybe the platform can help people to validate it, but then the use of the platform is, gets you into those shorter, tighter cycles of validation on your process. So can you speak about that a little bit? 

Bob: Yeah, let me, let me use the, uh, the Tulip example here. When we implemented Tulip at AstraZeneca. I'm not going to stop you.

So we don't bring anybody else in. Yeah. Yeah. It's just you and I, right? So, so we bring in Tulip and Tulip is a platform. Here it is. Okay. And so now we're going to go validate it. And I'm not involved because I'm worrying about other things, the global level and stuff. And they want to bring it in and eventually we'll look at it.

And they come and they say, so we're going to validate Tulip. I said, okay, so what does Tulip do? Now I know Tulip because I know you guys and I've seen you at conferences and different places and stuff and previous companies trying to come into it and everything. So I know what Tulip is, but I'm just being a wise guy.

So what are we going to do? Well, Tulip, you can do a lot with it. Okay. So what can you do with it? Well, you can build stuff. Okay. So what do you need me for? Well, we want to validate it. Well, validate what? We want to validate the platform. What does that mean? Validate the platform. Well, we're going to qualify it then.

Well, what does that mean? We're going to make sure that we can install it and it works. But what have you put into it? Nothing. So Tulip, we have to build applications, right? We can build, let's say any app we want, and that sits on the Tulip platform. So if you have no apps on it, what's the intended use?

And people go back and now they're writing validation plans and specifications on the platform and coming back and telling us the platform has an audit trail and everything's wonderful, but what did that do for us? What did that do for us? And nobody knows. I don't know. 

Natan: Yeah. 

Bob: Right. I still say so. Okay.

So now what you're telling me is, is I have an application now I want to build on Tulip. What do I do with the application? You can use it on Tulip. Yeah, but I have to configure it and build the whole application now. So what do I need to do? Oh, we don't know that. Well, that's what we want to focus on.

That's what matters, right? Let's set up those things and let's look at it. And the difference is between a platform and a application because you get other examples like, I don't know if I'm making sense here, but you get Power BI, Power BI as a platform. Yeah, same thing. I can't do anything with Power BI.

I need to ingest data somewhere so I can pull data from it. And then I need to build a report or a dashboard or something to use it. So what does the platform do for me? Nothing, but we've qualified it and we've done this and we've told you you can do this. And what did you do? Well, we installed the software on a server.

Okay. How did you do it? Well, we followed what the vendor, what they told us. Tulip told us to do it this way. That's what we did. Okay, good. Move on. And we get all caught up with that stuff. 

Natan: How did that translate into the actual adoption? You know, because what was interesting in the Aster case is that the use cases got validated in a more mechanized fashion, what you all call lovingly TLS.

Can you explain a little bit about TLS and how did it address this issue? 

Bob: So this becomes the challenge, right? It's like, okay, so Tulip has this piece of software. They have a platform and our folks are saying we've qualified the platform. Great. Now we have nothing, but we've done all this work. It's taken us all this time and we can't, still can't use it.

So how do we use it? Well, we have to put these applications in. Okay. So how are we going to validate the applications? We're going to use our existing processing. So again, the existing process, the same way we did the platform. Yes. So every time we write an application, we're going to write validation plans, requirement documents, design specs, risk assessments.

Okay. So what do you want from me? Well, the sites and people are complaining about it. Why are they complaining about it? Because it's going to take too long. Well, that's not my problem. Of course it's going to take too long. And when you look at it, you sit back and you say, what are we trying to do with Tulip?

We want to just grow into it. We want to digitize a piece of paper. Okay. We want to take a training room roster, you know, log into a training class where it was on paper and we want to digitize that. Simple. Yeah. Okay. For argument's sake, it's a GMP document. Let's not, let's not pick hair. So let's just say it's a GMP document.

So if I want to digitize that, why am I writing a validation plan, a requirement document, a design spec, a functional risk assessment, testing, testing, testing, testing. Because that's 

Natan: the process, Bob. Of course, that's the reason you do it. Because that's the process, Natan. Right. What kind of question is that?

Bob: See, you should be in quality. I am in quality. That's right. That's the process, Bob. That's what you need to follow. It's not flexible. And then, Natan, you and I will go out for lunch, you being in quality, and you'll tell me, Bob, you should actually take a risk based approach on validation. That's right. But that's what I was trying to do, Natan, is take the risk based approach and say, don't do all that stuff.

Don't do it all. Just take the form. That's your requirement. Go configure it. Do a quick run through, document three steps that you ran through, it didn't work, and we're done. Amazing. Exactly. So what we did was we, we sat out to do this. I said, well, listen, since everybody's calling me and want, want some help on this, let's sit down.

And, and so I'm on the West coast in the US. So my day starts at about 4am, 4. 30 in the morning, because AstraZeneca is worldwide and majority of the folks are over in Europe. What could then. I have time in the afternoon, my time to work with folks from AsiaPAC because it's their morning, my afternoon. So it seems like I work around the clock because in the evening time, we use one of your, your trusted partners, their factory talk to help us.

Implement Tulip at AstraZeneca. And so I met with the folks at FactoryTalk, and said, listen, why don't we use Tulip to help us validate Tulip? MetaTulip. So if Tulip is that good, then let's figure it out right now and decide, is it that good? So my ask to FactoryTalk was, I will work every night with you guys.

For as long as it takes to build a validation process using Tulip. So that'd be one application, 10 applications, 20 applications to build a solution. We're going to see how great Tulip actually is. We're going to see how fast we can do this and that will be the use case and the solution, the deciding factor, should we continue to use Tulip or not?

So that Pippin and I went. And we spent many, many nights, a lot of weeks over a whole summer, a couple of years ago, and we came out with what we called Tulip Lifecycle Solution. And it starts, I think there's about 12 applications in it. Starts with the demand. Somebody has a demand, they log it in, it asks some questions, you answer the questions, it decides how it's GMP or not, and all kinds of stuff, and what it does is, in the end, it gives you a model.

It tells you you're, I think we were calling categories at the time, you're category one. Your category 2, 3, 4, in each of those categories and models has a different validation life cycle. So they're not all done the same and all of them are done within Tulip. So if we need to write requirements, you'll write them in Tulip.

There's a little app to write requirements. There's a little app to write a risk assessment. There's one to capture test scripts. All this stuff is in there, the entire end to end life cycle. 

Natan: Beautiful. And you're out of documents and SOPs and? 

Bob: There's one SOP that runs the whole thing. 

Natan: By 

Bob: the way, the SOP?

is an SOP in our document system. It's also all instructed in the tool. So if you hover over something, it's going to tell you what you need to do anyway. And there's a diagram for it. So we have it both ways and we have all that stuff there. 

Natan: And then you have consistent, every interaction with any step of the way.

Documented, and timestamped, and e signed, 

Bob: e signed where needed, all consistency from cradle to grave. So when you, when you made the decision it was GMP, it's all right there. You can see why you did it. You can run a list of every app, all right there. And what we did was, we have at Ashizeneca, many sites around the world.

So every site has their own Tulip instance. And then we build a global one. We put this one on the global one. So everybody has to tap into the global one. So everybody follows the same process and then you have your local systems. So we went through it and we got it done a couple of years ago. And I told Don and Pippin, I said, you know, Philip's way ahead of their time.

This is fantastic. Look what we did. Look what we did. We built 12 Tulip apps for an entire life cycle. 

Natan: Yeah. 

Bob: Think about what we're replacing. 

Natan: A lot of systems, I think, a lot of paper, a lot of fragmentation of the processes actually gets adopted in between sites. A lot of training. 

Bob: And there's other systems out there that do this.

Right? You can buy point solutions to do these things. And look what we did. We built it in just several months. So to me, it goes back and says, Tulip can do what they say they do. You can build apps quickly. And it worked. 

Natan: I saw some bits and pieces of this, of course, talk to you all along the way, because I think the work you're doing is super exciting.

And we're just scratching the surface and, you know, in the true spirit of an open operations ecosystem community through. Regulated Industries and others, like we're learning from each other and, you know, I've seen different things in some of our companies that are more in the dual use, aerospace defense and they have different type of like EDHR like type stuff and so I'm learning, I'm learning from what you all did and what other folks have been doing.

Maybe you would call what they're doing, like, we say EDHR, but is it really MRO? And so like, there's like this blur of, uh, different ways of doing things. And like this notion of MetaTulip that suddenly allows you, basically what you're describing to me is like, I knew how I wanted to change how I worked.

I made it happen with the platform. And I think like, separate that from Tulip for a second. Cause like the whole point of frontline operation platforms and human centric Industrial software, that kind of stuff is that the people who know the work can actually take control and decide how they're going to fashion technology, not the other way around.

And in that case, that's what I love about the story. And I think that is just the beginning. And I think here in this story, it informs How you deal with a very complicated subject of, you know, rolling up something like, uh, you know, no codes is in development in pharmaceuticals, which is, uh, you know, if you would ask people like a few years ago, I don't know if it's three or five or something like that, they would tell you like, listen, we're still trying to figure out this, uh, cloud thing.

What are you talking about? No code. And, and now you're seeing thousands of users, not just Astro, but across so many of Tier ones we're working with and that is just exciting. And you know why this is happening? I'll tell you, because in all these companies, there are people who know what the work is and they need the tools.

And if they can get platforms that can do that, then they will decide within the constraint and here the constraint is quality and regulation. what to do. Otherwise they won't do it. And I think that's more enduring, but Bob, thank you so much for coming on the show and joining us today. 

Announcer: My pleasure. Thank you for listening to the Augmented Ops podcast from Tulip Interfaces.

We hope you found this week's episode informative and inspiring. You can find the show on LinkedIn and YouTube or at tulip. co slash podcast. If you enjoyed this episode, please leave us a rating or review on iTunes or wherever you listen to your podcasts. Until next time!