Software Quality Today

Digitizing and Integrating the Validation Ecosystem, with Phil Jarvis

Dori Gonzalez-Acevedo

We're excited to welcome Director of Integrated C&Q and Kneat Services at VEQTOR, Phil Jarvis! Phil is extremely active in the life sciences community and is a staunch advocate for modernization and improvement in our industry backed by a wealth of experience and insights.

I have to apologize in advance for the audio. We had a bit of a technical issue when recording this live, but trust me, this conversation is well worth a listen as Dori and Phil dive deep on the importance of digitizing and integrating the entire validation ecosystem.

Please follow Phil on LinkedIn here and don't forget to 

*Disclaimer: Podcast guest participated in the podcast as an individual subject matter expert and contributor. The views and opinions they share are not necessarily shared by their employer. Nor should any reference to specific products or services be interpreted as commercial endorsements by their current employer.

This is a production of ProcellaRX

Dori Gonzalez-Acevedo:

Well, welcome to another episode of software quality. Today, I'm super excited to be able to interview Phil Jarvis here live at Connects validation University. We're in Philadelphia and Phil, you're in from the, in from Ireland in from Ireland from sunny, sunny Limerick. Left, and it was raining here yesterday. So that wasn't really fun either.

Unknown:

It wasn't much of a climatic shock to me either, which, which is good. But no, the travel was good. And he's delighted to be here with you today.

Dori Gonzalez-Acevedo:

Yeah, great. Thanks. I know you were in December you were with connects live when we were over in Ireland, in Dublin, Dublin. And so you guys have been joining now as new sponsors for connects

Unknown:

we have I've had a long standing affiliation with John, ever since he left IVT and started connects. So I think I've been faculty since it started. But no, I really enjoy these events just because of the connectivity and the knowledge sharing and the actual real life examples that are shared, I see real value in it. And I think as part of factor now, you know, we see ourselves sponsoring and growing more with that partnership, getting involved in publications also, with connects, and also maybe on the training side as well in the future as well.

Dori Gonzalez-Acevedo:

That's awesome. Well, why don't we start with telling us a little bit about how you came to be in validation? Because it's not something that we all aspire to be?

Unknown:

No, no, I don't think you ever sort of go I'm going to be a validation engineer. No. I started off being a chemist. So I have two best people. And then went on to the lab from being a chemist, and then obviously got into sort of HPLC, validation, etc. Because no one else wanted to do it.

Dori Gonzalez-Acevedo:

With G factor is de facto role we had to play. Yeah.

Unknown:

So from there got into more manufacturing, equipment validation, process validation. I worked in j&j for about 10 years, and various validation leads sorts of jobs. Until or, I would say that was about 1015 years ago now. And then I went on to a big bio startup in in Cork BioMarin. Oh, by marine, yeah, which was their validation lead desk. That was my first step of Greenfield sites. So that was all exciting. And then really got into one of my passions there, which was more around risk based validation. So how can we do things more efficient? How can cause Bahrain is rare diseases. So we were really pushed with a mission of this is the only medicine that is out there for this disease. And one of the great things that Iran did is they had a treatment for battens disease, nothing out there. And I don't know if you know anything, no, I don't think so effects, usually children, they don't usually live past, you know, 40, because the thing is very debilitating disease, nothing out there to treat it at the moment. So you know, very much like trying to get that treatment out there as fast as possible, embracing the most lean methodologies available out there. And through that, that led me to, well, if we can only go so far with paper, what can we do when we digitalize things? And that has been my passion property for the last six to seven years now? How can we digitalize these processes? How can we link systems together so we can create this digital ecosystem and share data and become much more data centric than paper centric? Yeah,

Dori Gonzalez-Acevedo:

I know you and I share very much the same sort of mindset around that. And it's, and then you were at Abbey, as well, I was

Unknown:

Yeah. So I went from Bahrain, and I went to AFI, I was there as the EU, seeing Q lead for three years. And that was really developing both their risk based CQ Strategy off the ISP baseline guide version two. So my boss, Laurie Kim, at the time, he was a very, you know, she was a great mentor to me very. I would say inspirational woman. Okay. She was one of your co authors of the ISP guides, and she said, look, let's get ahead of the curve. I want to push this into Byron and Amin heard that. But then at the same time, we said let's digitize this at the same time. We rolled out in the last three or four years, we've rolled out to I think, eight sites within the ERP network.

Dori Gonzalez-Acevedo:

Yeah, that's awesome. And then you bridge the gap and you've jumped over to the dark side on the consultant inside.

Unknown:

Yeah, so I don't know, have you ever met Martin rear row Deki. From vector he's now directors to his very inspirational sorts of full of energy.

Dori Gonzalez-Acevedo:

I have to meet him, I love him.

Unknown:

But he's just a ball of energy. So vector was the partner for HUD V, rolling out neat. And, you know, had really good experience with them as a consultant helping me at the time. And I was beginning to evaluate, like, how can I influence wider than than then that's just one organization, and Vivian Fenton. So another woman in pharma that I admire, amazing, amazing woman, she's, you know, she's in the automation space, very rare. Yeah, someone in that space? And she was like, would you consider coming over to, you know, the consultancy side? And really, it's been a revelation to me, because not only do you have spare time to go to conferences like this. I'm very active in the ISP, I'm active in Irish universities and upskilling validation staff in in Ireland, but also is great work on like, a very big variety of projects. So no, it's been, it's been very good so far, but very different.

Dori Gonzalez-Acevedo:

Yeah, different. And I like, for me, too, is the variety and be able to spread that wealth of knowledge across a much, much more broader spectrum and really helps inspire more, right? And then how do we tap into that next generation, just like you're saying, You're doing the Irish universities, right? Like, how can we teach this stuff? Because we were never taught this? You would

Unknown:

never talk? Yes, exactly. And that's one of my great passion. So about what I think is about six or seven years ago, like islands just exploding by attack wise moment, and there was just a complete shortage of validations. That staff. So we got together a lot of companies with IBEC, which is one of the Irish government all sorts of authorities there and said, How do we increase the visibility of validation? Being a great career? Yeah, because people were just not aware, though, is out there. And said, right, well, we'll start small. And we will do some upskilling courses, some Springboard courses, they call them. And they pulled experts from the Irish validation community, and created very interactive sorts of classes with case studies like that. So we got really good feedback from there, but and now they're doing masters courses, modules for BSC honors courses in biotech. So it's great because we get people now that have a solid understanding of validation coming out of university. The other thing is, and this is a bit of a spoiler alert, actually, me and neat were Vectra an official partner with NEET. We are working on Irish universities validation instance of meat, so that they will use need as part of digital validation modules going forward. Yeah. So

Dori Gonzalez-Acevedo:

that's a great, yeah. So how to incorporate these things from the very beginning, rather than waiting to folks getting into a career or into a full time position at a big organization to have that experience, but actually get it in. That's, that's great. That's awesome.

Unknown:

I mean, he, you know, I think it really gives benefit to all sorts of parties, because we get staff rates, right, you can hire Yep. But now the skill, the skill sets, but also, we also have an awareness right from them that this is a career and what skill sets that

Dori Gonzalez-Acevedo:

what's needed. And so actually going to be launched tomorrow on the podcast is my episode with Sue black from the UK. She is doing similar in her work on tech up women were really partnering with companies to be very specific about what the skills are, that are needed, and then teach those skills because they know what they are. We just haven't educated across the board for what they need to be.

Unknown:

And I think those skill sets are constantly evolving, especially your way technology is evolving so fast. Yeah. Like critical thinking, decision making. You know, creating good knowledge management process, etc. Yeah. So it's yeah, it's great to be involved, something that I'm highly passionate about, like trying to teach you know, you know, give give young people a great start talking b is an amazing career.

Dori Gonzalez-Acevedo:

Yeah, that sounds so the one of the things that I'm interested in and I was wondering, pick your brain about see what you think you know about because you're often in the commissioning side and equipment commissioning side, there's so much new technology coming out there with how to do literally validation on the floor, right, which is amazing and awesome. And yet I can hear in the back of my head, some quality folk may be perhaps hesitant on wanting to adopt some of that new technology. So do you have any thoughts about that?

Unknown:

Yeah, so it's something you know, we definitely have had to contend with with with multiple clients at the moment because a lot of especially going to digital validation practices or very new technologies there there is that hesitancy there? Because I think we're we're a very conservative industry at the end of the day. But I, I think there's there's more of a drive by companies now of the speed to market. I think that's really helping pull the quality folks into the fold, and become involved in so sort of solving that problem. And a lot of these things are big cultural changes as well. They're not easy to

Dori Gonzalez-Acevedo:

do. Right? Yeah, cultural changes, which include process change, which includes people change, right. And I think you and I have talked about this, it's not the tools we can do the tool part is the easy part, right? It's the people part that we have to help along the way how to enable how to utilize the systems that were putting into place the most effectively as that they're designed for, right? Like, a lot of these tools do a lot of things, but not everyone uses all of those things.

Unknown:

How would you provide people with really top class training? A lot of our clients at the moment, especially in the need space, you look at them wanting to make their training more efficient. So looking at more CBT types of courses. But a lot of the time, there's so many scenarios in validation that you can't be an instructor led sort of classroom based session as well. But I think you raise a really, really good point. There's the new technologies and new tools that are coming out. But especially in the digital space, there are huge challenges of how we're going to validate things. So if we look at AI, deep learning, machine learning, how do we actually validate something those determinating its own decisions? Yeah, yeah. Yeah. And these are really exciting challenges, right, that we need to solve quite quickly. Yeah.

Dori Gonzalez-Acevedo:

And have to think about differently. Right. So getting back to your earlier point about critical thinking, right? Like, we can't take the same methodologies that we've been doing for the last 1520 years, right and apply it to the same way, it doesn't work that way. It's not possible. Which leads me to more of the agile mindset conversations, right? And how to help organizations make that transition, just even thinking about that in a different way. Right.

Unknown:

And I have a case study, I have a session tomorrow at Connects around farmer 4.0. And there's an example of that in machine learning. So how do you get people comfortable, especially, especially the quality organization? So I use an example in there a vision system? Okay. So let's say you're inspecting a medical device, let's say it's a contact lens. Okay. A lot of the time, you are teaching that vision system, what is good and what's bad, if you go into something that is using machine learning, is going to begin determining what's good and what's bad. And that's quite scary for quality. So what can you put in place the sort of the pieces that that can get the quality mindset around? Okay, is it doing a good job? Is it making good decision? Well, you can put parallel activities in place, so maybe, okay, you allow it to make that decision. But then that's vetted by a human being, yes, that is bad, a new talent, talent is bad. Or maybe for a certain amount of time, you will have maybe a QC check that's ongoing with. So I think there's ways of beginning to make people

Dori Gonzalez-Acevedo:

comfortable the approach. Yeah, I think it's similar. We saw this in software quality testing and test automation, right, folks really not understood or not understanding the technology about test automation, right, and having to do that parallel sort of approach for a while, right, running a manual test and then running an automated test and showing that it's the same thing. Now where we're at with the automation tools from a test software quality test. You don't need to do that anymore. Right. But we didn't need to do that in the beginning maybe. Granted, some folks are still doing that. But the technology has well surpass the the need to do that pair Allow approval, if you will, from test automation perspective and software quality testing, be the same for machine learning and other types of where you physically can do that visual inspection versus maybe something else to do that comparison for us, right.

Unknown:

Yeah, exactly. And it's just, it's helping that process and along its way, just getting people comfortable with that route challenge that.

Dori Gonzalez-Acevedo:

Yeah. The other thing I was wondering, you know, this. So, again, I often come from a software development kind of perspective, but the constant change that's needed in order to get out there today, right? It's, we're ideating as we go. But that's the same on the production floor. Yep. Yeah. Right. So I mean, in that, so when I, back in the day when I was on pilot and doing chemical process engineering, right, we did hundreds of of experiments in tort until we got to the final one, right. So what do you think about it? Has anything really fundamentally changed? Or is it Are we are we able to adopt a, that iterative process on so we were talking about validation? 4.0 and pharma performed pretty well, right? So really continuous manufacturing chain, right, which is different than 15 years ago, where it was a discrete set tasks? Does that make sense?

Unknown:

Indeed, and I think a lot of that comes down to a lot of what's been discussed here today, by some good speakers, like, what is our decision making process? What is the risk? So you know, we have to say, well, if I'm making something those a treatment for a year person, especially in the ATMP, sort of space, where it is literally the only treatment with someone that's dying, or what is the risk of

Dori Gonzalez-Acevedo:

greater they're gonna die or not. Anyway, right. Yeah, exactly. Yeah.

Unknown:

So that some of that is around the risk and the decision making, right when talking to talking about, and then some of it needs to be, you know, so So what's so bad about failing? Great question about

Dori Gonzalez-Acevedo:

learning. Yeah. So that's giving up the being perfectionist, which I'll talk about tomorrow, and my authentic leadership, that sort of thing, right is giving up that we always have to get it right all the time, right. And as part of the quality mindset that perhaps needs to be set aside and failing first failing fast is was is regular for like software development, but we need to actually embed that into everything. Right.

Unknown:

But as you say, is fundamental to the whole production process. When we do pharmaceutical design. Yeah, as per ich, QQ Q eight, you know, we're there right up again, during the world do we use finding our edge? Yeah, no, you shouldn't just stop there, right now. informs the process process going forward. But I think we just need to be more cognizant of risk. Yeah, more things. Actually.

Dori Gonzalez-Acevedo:

I thought it was interesting this morning. What James Mason was talking about the hand sanitizers, did you did you hear that? Oh, you didn't see that part. So the FDA saw that the the in the warning letters and things around the hand sanitizers went up because everyone was trying to get hand sanitizers out during the vaccines, right. But there was a lot of knockoffs that was being done and not being tested properly. And it just highlights the fact that risk constantly has to be reevaluated. Right? Because it's something like that we couldn't have anticipated before. But then it's just a very good way to say, yeah, it's a constant looking at risk, right? So not just doing a risk assessment from the very one time it's done. It's not that it's this constant living sort of breathing thing. One of my pet peeves around requirements, specifications, right, is that, you know, folks want to take this requirements specification and put it in a dock management system and think that it's done. Requirements are living, right. We all you know, designers and engineers understand that, like that's constantly evolving. And part of that is looking at the risk each and every time that we're doing something.

Unknown:

Yeah. And I think digitization is going to help us out as well, because we're going to have much more advanced process controls in place, especially if where companies are, you know, definitely, you know, we saw in one of the ISP polls at that moment, that there was done recently there was a survey by the CQ cop, and more and more companies are looking towards digitalization and I think that helps with the constant assessment risk, because you're able to use pap technologies to say, or this attribute isn't quite right at this step, even if it is a continuous process. What can we do maybe then to have just that process. And maybe that's where machine learning comes in. If we can build up all these technologies on top of each other, we get a much more dynamic and a much stronger control strategy that then allows us to make

Dori Gonzalez-Acevedo:

good decisions. Yeah. And along those lines, I didn't know I could imagine. And when you're you're implementing NEET for a customer. Similar is you're, you're looking at their business process to kind of align some of those data points, right, so that they're not totally whatever they think, right? You're trying to help them guide them towards standardizing their data to not just implementing a system, right? Because it's more than just implementing a system, it's actually standardizing their data so that they can make meaning of that data as they go along. Because if we don't do that process, right, if we don't go through and have that standardization, we can't make any further predictions or analysis. Right. Yeah.

Unknown:

And I think that's, that's a great challenge for the future as well. How do you get all of these systems talking to each other in a standard language without this need for middleware to convert?

Dori Gonzalez-Acevedo:

Yeah, I know, we were talking about this on the the ISP subcommittee that we're on, right, like that is like the ideal, right, like how to do that. And part of my, you know, lessons learned over the last 15 years is is, you know, companies ask for lots of things. But as a consultant, they might not be asking for the right thing. And there's a way that we can to given that we see a lot and a lot of different varieties out there. Being consultants, we can actually help standardize the industry better if we do that collectively. And so that's one of the things I love about the ISP subcommittees as well as in you know, connects like this, right, we work collaborate on what are the best industry practices? What do we want to move forward? One of the things I was thinking about when you were sharing also with the next generation coming in? is data science in general. Any thoughts about that?

Unknown:

Yeah, I think data analytics and data sciences, this is going to become more and more of a growth need within the industry. Which is why, you know, it's great to get the things like meat in Irish universities level. So they know that these the systems are capable of doing these things. We have a very good partnership, actually, with a data analytics company as part of vector. So we see that strategically being like definitely a growth sort of thing within vector as well. But yeah, making sense of data content, contextualizing the data, and making decisions of that data, and improving, you know, is this the right the right data? Because statistics can lie.

Dori Gonzalez-Acevedo:

Bridge rates? Exactly. Yeah, you can pick and choose poorly.

Unknown:

I've had very blended career, I've been very blessed that, I guess that way, I spend half of my life in medical devices, which is very statistically based, yeah. And then a lot of time in biotech, and then steroids, etc. So, you know, having that statistical base, as you say, you get to challenge the data, you know, so, you know, if I have data, which looks great, but it's not normal, you know, yeah. What's that really telling? Yeah, that's right. You've got to have some, you know, general presumptions about that data. Yeah, or assessment of the data. So I think data's gonna become a huge thing in the future, especially when we're looking at things like real time release, especially with things like technologies like ATMP is when you're going to want that real time release to give it to a patient.

Dori Gonzalez-Acevedo:

Yeah. Yeah. Awesome. Are there other things that you want to touch on while we have time today? Plans for the future things that things that you

Unknown:

plans for the future? I think, you know, I'm learning a lot. I think the great thing about the consultancy is that you're learning something new every single day at the moment. I'm really excited about the future, the partnership with neat, especially as their new version 9.0 goes on. And there's much more powerful, powerful functionality to connect to other systems and create that digital weak eco structure that we're talking about. But really excited to be at the event today. And actually, you know, talking to people

Dori Gonzalez-Acevedo:

I know. Yeah, it was. I was at s QA a couple of weeks ago, and it was, like 800 people, and it was a lot of people came out because it was the first time was full in person and really enjoying in community. Right. And one of the things I think you said earlier around the connects community was we're trying to build more support structure for everyone. For us, you know, wherever you're at within your career, right, so folks like yourself, and I can come in and help guide folks from the very beginning, give them some questions space to ask questions. And, and then with the FDA here as well, we've gotten great representation from from the FDA. So

Unknown:

the thing that I would say is I'm not I'm very positive about the future as well, because, you know, we're hiring a lot of new engineers to support our growth in Ireland. And what we're finding is we're recruiting especially at a junior level, definitely people younger generation or myself, yes. Yeah. And, you know, but they're so bright, especially when it comes to digital systems. Yeah. They're teaching me things. Yeah.

Dori Gonzalez-Acevedo:

They're digital natives. Right. And so that's where they really grew up. And I am I'm, the ideas that they have are amazing, right. And so my hope is to be able to remind folks of our generation to listen to them.

Unknown:

You can learn amazing things off. Yes. Yeah, absolutely. Well, thanks, Phil, I

Dori Gonzalez-Acevedo:

appreciate your time today.

Unknown:

I really appreciate you for having me. And it's great to see you in person, rather than just Oh, yeah,

Dori Gonzalez-Acevedo:

it's absolutely. Alright. Well, I have to come over to Ireland soon.

Unknown:

Oh, definitely. Invited. Yeah, it's locked in. As long as you like Guinness and you like the rain.

Dori Gonzalez-Acevedo:

It's good. It's very good. All right. Well, back to the show here. Thanks.

People on this episode