1

00:00:10,575 --> 00:00:11,135

Brandy: Hi everyone.

2

00:00:11,255 --> 00:00:15,197

Brandy: Today I'm thrilled to welcome Ramila Peiris to the podcast.

3

00:00:15,257 --> 00:00:24,941

Brandy: Ramila is a pragmatic and strategic leader with over a decade of experience driving innovation and delivering impactful data science solutions in the pharmaceutical,

4

00:00:26,354 --> 00:00:27,955

Brandy: and biotechnology sectors.

5

00:00:28,535 --> 00:00:36,439

Brandy: He's currently serving as the global head of data management, ML and AI platform within MSAT at Sanofi.

6

00:00:37,099 --> 00:00:44,522

Brandy: He has dedicated his career to transforming business processes, supporting the adoption of AI and building high-performance teams.

7

00:00:45,242 --> 00:00:54,687

Brandy: Ramillo's passion lies in leveraging data to make faster, more informed decisions and creating practical solutions that address critical challenges in biopharma.

8

00:00:55,507 --> 00:01:03,955

Brandy: He believes in the power of storytelling to gain buy-in from stakeholders and sees data as a foundation for game-changing advancements in the industry.

9

00:01:04,916 --> 00:01:14,644

Brandy: Today, we will dive into his journey, his perspective on AI and data and biopharma, and his vision for the future of manufacturing digitization.

10

00:01:14,664 --> 00:01:17,166

Brandy: Ramila, welcome to the podcast.

11

00:01:18,568 --> 00:01:19,168

Ramila Peiris: Thank you, Brandy.

12

00:01:19,188 --> 00:01:19,909

Ramila Peiris: Thank you for having me.

13

00:01:20,868 --> 00:01:21,970

Brandy: Yes, we're very excited.

14

00:01:22,030 --> 00:01:23,994

Brandy: I think we're just going to jump right in here.

15

00:01:24,194 --> 00:01:32,329

Brandy: And how about you share a little bit about your background and what led you to focus on data-driven solutions specifically in the biopharma industry?

16

00:01:33,939 --> 00:01:34,880

Ramila Peiris: Yeah, sure.

17

00:01:35,060 --> 00:01:35,260

Ramila Peiris: Yes.

18

00:01:35,340 --> 00:01:39,462

Ramila Peiris: So my background is in chemical and process engineering.

19

00:01:39,482 --> 00:01:46,806

Ramila Peiris: I specialized in mathematical modeling, mechanistic modeling in during my graduate studies.

20

00:01:48,147 --> 00:01:52,109

Ramila Peiris: I think that provided me a solid background to work in data science.

21

00:01:53,110 --> 00:02:00,337

Ramila Peiris: When I first joined the pharma industry, I was fortunate enough to be thrown into process troubleshooting.

22

00:02:01,758 --> 00:02:08,645

Ramila Peiris: With data analytics, I saw the impact that it made, and I love that experience.

23

00:02:10,106 --> 00:02:25,776

Ramila Peiris: I also realized that my process engineering background really helped me to think through problems and apply analytics in a way that is meaningful to experts that I was interacting with and helping.

24

00:02:27,822 --> 00:02:34,611

Brandy: Yeah, I could definitely see how all those worlds converge to make your role very impactful.

25

00:02:34,631 --> 00:02:43,823

Brandy: You know, kind of thinking about what drives your passion in championing the role of data and making these impactful decisions within pharma.

26

00:02:44,003 --> 00:02:45,485

Brandy: What is it that drives you?

27

00:02:47,352 --> 00:02:54,020

Ramila Peiris: You know, the pharmaceutical industry impacts public health in a positive way, right?

28

00:02:54,861 --> 00:03:01,569

Ramila Peiris: So I believe data is truly a key enabler for problem solving in pharma industry.

29

00:03:03,151 --> 00:03:06,776

Ramila Peiris: So that belief drives me to champion the role of data.

30

00:03:07,416 --> 00:03:24,754

Ramila Peiris: work on initiatives that make data ready to use and develop data analytics solutions that help solve problems faster, whether it is process troubleshooting, improving manufacturing processes, or improving business processes.

31

00:03:26,744 --> 00:03:36,154

Brandy: Yeah, and part of that too is right, transforming speed and cost of delivery to customers and especially in the pharma world.

32

00:03:37,716 --> 00:03:45,064

Brandy: Just kind of wondering if you might be able to elaborate a little bit further and give some examples of how this works in practice.

33

00:03:47,659 --> 00:03:52,181

Ramila Peiris: Yeah, in our industry, the quality of the product is very important.

34

00:03:53,582 --> 00:04:02,146

Ramila Peiris: Now, if you have a product quality issue, we may end up in a situation that we are not able to meet the market demand.

35

00:04:02,786 --> 00:04:05,908

Ramila Peiris: And hence, we may face the risk of losing market share.

36

00:04:07,008 --> 00:04:14,592

Ramila Peiris: So solving that product quality issue as fast as possible is extremely important for us.

37

00:04:16,913 --> 00:04:25,337

Ramila Peiris: And also the same thing is true if you have a yield issue where the manufacturing output is below the expected target.

38

00:04:26,498 --> 00:04:34,382

Ramila Peiris: So when we face situations like that, we spend time and efforts in solving these problems.

39

00:04:35,162 --> 00:04:37,183

Ramila Peiris: And sometimes a lot of experts are involved.

40

00:04:37,984 --> 00:04:40,325

Ramila Peiris: That means they have their time and their efforts.

41

00:04:41,646 --> 00:04:46,449

Ramila Peiris: So these experts use data to make decisions to solve problems.

42

00:04:47,149 --> 00:05:05,321

Ramila Peiris: So if we can provide data in a ready-to-use format and analytics in a ready-to-use format, what I call analytics at speed, the problem-solving times can be reduced and thereby reducing cost of manufacturing.

43

00:05:07,318 --> 00:05:27,397

Brandy: Yeah, and I would imagine just with the addition of AI into this world, I would imagine that there are a lot of other obstacles that can be solved in a more timely manner.

44

00:05:27,797 --> 00:05:32,061

Brandy: How are you kind of seeing AI play into the overall

45

00:05:33,920 --> 00:05:37,223

Brandy: reduction of obstacles in the manufacturing environment?

46

00:05:37,243 --> 00:05:54,280

Ramila Peiris: Yeah, I mean, AI, we call it AI, AI or simple data analysis, regardless of the methodologies we use, it feeds into decision making at the end.

47

00:05:55,084 --> 00:06:19,115

Ramila Peiris: And the way I think about is any analytics or AI, whether it is AI or simple statistics, how can we deliver that capability faster in a meaningful way so that our SMEs and subject matter experts don't waste their time and they can have that capability readily available.

48

00:06:21,810 --> 00:06:23,652

Ramila Peiris: you know, we can do lots of things with AI.

49

00:06:24,033 --> 00:06:32,963

Ramila Peiris: And for me, the true capability of AI is, you know, how can we deliver problem-solving capability through AI faster?

50

00:06:34,727 --> 00:06:42,352

Brandy: Yeah, and something that you've mentioned before is there's no AI without information architecture, without IA.

51

00:06:44,714 --> 00:06:53,460

Brandy: You know, could you elaborate on this and explain why information architecture is really the foundation to effectively leveraging AI?

52

00:06:56,373 --> 00:06:57,534

Ramila Peiris: Yeah, this is very true.

53

00:06:57,554 --> 00:07:06,418

Ramila Peiris: In my opinion, the main reason is that many industries are not tackling the right problem when it's come to AI.

54

00:07:06,438 --> 00:07:07,959

Ramila Peiris: You need the right

55

00:07:11,520 --> 00:07:18,251

Ramila Peiris: kind of information architect to have AI at scale.

56

00:07:18,792 --> 00:07:26,304

Ramila Peiris: What do I mean by that is, you can build up solutions, but the sustainable use of that solution is important, right?

57

00:07:26,804 --> 00:07:41,308

Ramila Peiris: So most of the time we are trying to bring solutions, AI solutions, which normally AI is a buzzword and it's shiny.

58

00:07:41,828 --> 00:07:48,070

Ramila Peiris: And we are bringing these solutions without thinking about how can we deliver the capability in a sustainable way.

59

00:07:49,050 --> 00:07:53,671

Ramila Peiris: And that's where the data information architecture is so important.

60

00:07:56,086 --> 00:07:57,506

Ramila Peiris: Let me give you an example.

61

00:07:58,187 --> 00:08:05,050

Ramila Peiris: If you buy a house, you expect the faucets in your house to work.

62

00:08:05,770 --> 00:08:20,697

Ramila Peiris: There is an unconscious expectation that the house you walk in, when you buy a house, will come with pipelines to bring water to your bathrooms, kitchens.

63

00:08:22,029 --> 00:08:28,895

Ramila Peiris: Typically, these water pipelines are designed and installed during the construction of a house.

64

00:08:30,456 --> 00:08:33,439

Ramila Peiris: Now think about a pharmaceutical manufacturing building.

65

00:08:34,020 --> 00:08:46,031

Ramila Peiris: Do we consider creating data systems that clean, organize, contextualize data during the planning of construction stages of a new building?

66

00:08:47,471 --> 00:08:48,992

Ramila Peiris: Sadly, most of the time we don't.

67

00:08:50,893 --> 00:09:07,820

Ramila Peiris: So if you are building a state-of-the-art manufacturing facility that we expect to be ready for AI, we must also build foundational data capabilities, foundational system architecture that enables ready-to-use data analytics.

68

00:09:08,601 --> 00:09:13,283

Ramila Peiris: With that level of capabilities, we can create facilities that are AI ready,

69

00:09:14,303 --> 00:09:17,906

Ramila Peiris: where data is ready to use from day one.

70

00:09:19,987 --> 00:09:21,208

Brandy: I really like that analogy.

71

00:09:21,748 --> 00:09:22,329

Brandy: That was great.

72

00:09:22,809 --> 00:09:38,260

Brandy: I haven't really heard anybody explain it quite like that with the pipes and faucets to a house, but do you think that that is one of the primary barriers to a lot of organizations not being able to fully realize the benefits of

73

00:09:39,474 --> 00:09:49,962

Brandy: the buzzword AI effectively is just because the architecture, the information just, they didn't build the foundation, the information to be able to fully leverage.

74

00:09:51,448 --> 00:09:55,011

Ramila Peiris: Yeah, I think so, because AI is tempting.

75

00:09:55,031 --> 00:10:00,355

Ramila Peiris: There is a buzzword, and there's a lot of promise that comes with AI.

76

00:10:02,356 --> 00:10:13,625

Ramila Peiris: Therefore, most of the time, we jump into creating AI solutions without really creating the capabilities that truly enable that AI solution.

77

00:10:15,686 --> 00:10:26,192

Ramila Peiris: So going back to my example, you got to build foundational capability in a house.

78

00:10:26,412 --> 00:10:33,696

Ramila Peiris: In my view, the ability to have ready to use water in a house is a foundational capability.

79

00:10:34,256 --> 00:10:37,459

Ramila Peiris: right, at least in the third world, in the developed world.

80

00:10:37,799 --> 00:10:54,194

Ramila Peiris: So I think most of it, there's a temptation to jump into AI topics without creating the information architecture in a proper way.

81

00:10:55,374 --> 00:11:04,456

Ramila Peiris: creating data systems, data pipelines, and taking that time to create more sustainable data architecture.

82

00:11:04,476 --> 00:11:17,738

Ramila Peiris: So most companies live with the promise of AI, but really don't achieve that because you don't have the foundation in place to have AI at scale.

83

00:11:19,338 --> 00:11:20,398

Brandy: Yeah, yeah.

84

00:11:20,498 --> 00:11:22,719

Brandy: So let's say that a company is

85

00:11:23,849 --> 00:11:38,621

Brandy: ready and prepared to kind of start to dabble in AI capabilities and, you know, Katalyze AI is focused on optimizing the bio manufacturing process, you know, using AI and data driven insights.

86

00:11:39,262 --> 00:11:48,990

Brandy: From your perspective, how could a platform like Katalyze AI help address some of the challenges that you've mentioned that the industry faces and

87

00:11:51,018 --> 00:11:56,467

Brandy: Like such as accelerating adoption of streamlining data integration and in pharmaceutical manufacturing.

88

00:11:56,527 --> 00:12:03,097

Brandy: Like how, how do you kind of see Katalyze a solution like Katalyze playing into this world?

89

00:12:05,011 --> 00:12:06,592

Ramila Peiris: Yeah, very good question.

90

00:12:06,712 --> 00:12:22,597

Ramila Peiris: I think when I think about AI and data-driven insights, I expect any good data and AI platform to solve the hardest part of data analytics, which is not analytics itself, by the way.

91

00:12:23,317 --> 00:12:29,979

Ramila Peiris: It is the ability to deliver contextualized and analytics-ready data to the end users and to the tools

92

00:12:32,647 --> 00:12:36,731

Ramila Peiris: Those companies often tout as game changing and groundbreaking.

93

00:12:37,832 --> 00:12:44,279

Ramila Peiris: Most platforms claim that they can solve the most difficult problems in healthcare.

94

00:12:45,040 --> 00:12:52,808

Ramila Peiris: If that is the case, I expect the platforms to solve the most fundamental problems that hinders us from doing analytics at speed.

95

00:12:54,489 --> 00:13:01,275

Ramila Peiris: which is creating contextualized data pipelines that are easier to maintain and easily scalable.

96

00:13:03,117 --> 00:13:11,344

Ramila Peiris: Any platform capable of doing that will also be capable of producing practical and pragmatic AI solutions.

97

00:13:13,726 --> 00:13:18,027

Brandy: Yeah, I mean, you nailed it right there, right?

98

00:13:18,447 --> 00:13:26,988

Brandy: The promises and hopes, but that is the piece that is really foundational to full realization.

99

00:13:28,268 --> 00:13:36,390

Brandy: Something that you had mentioned earlier and emphasized in your work is just the storytelling of data, right?

100

00:13:37,210 --> 00:13:42,531

Brandy: Getting buy-in from stakeholders, I have a feeling that you have a lot of experience with this over

101

00:13:43,832 --> 00:13:59,993

Brandy: over your journey in the, in the workplace and probably even in your studies, you know, could, could you just, how do you approach crafting stories that resonate with both technical and non-technical audiences?

102

00:14:00,093 --> 00:14:01,575

Brandy: Cause that is the true difficulty.

103

00:14:02,452 --> 00:14:07,815

Ramila Peiris: You know, I haven't really cracked the story telling problem yet.

104

00:14:08,035 --> 00:14:09,015

Ramila Peiris: I'm still learning.

105

00:14:09,816 --> 00:14:20,701

Ramila Peiris: What I would like to do is I like to come up with stories, examples that resonate with people, resonate with anyone.

106

00:14:22,141 --> 00:14:26,664

Ramila Peiris: If I can tell a story, even a child can understand.

107

00:14:26,764 --> 00:14:28,645

Ramila Peiris: And that's the message I want to give.

108

00:14:30,625 --> 00:14:38,329

Ramila Peiris: And then I believe if I can win over somebody who doesn't understand any technical parts, and that's the best story.

109

00:14:39,429 --> 00:14:44,451

Ramila Peiris: So especially with our leaders and decision makers, I try to do that.

110

00:14:46,012 --> 00:14:50,834

Ramila Peiris: I try to tell a simple story that is clear and understandable.

111

00:14:51,454 --> 00:14:58,457

Ramila Peiris: And then I like to try to explain a problem and solution with the simple examples

112

00:14:58,877 --> 00:15:14,945

Ramila Peiris: And most importantly, I like to tell the benefit of these solutions to try and win over the sponsors who can help us solve some of these data projects or some of these data problems.

113

00:15:16,400 --> 00:15:24,567

Brandy: Yeah, I mean, and this has been a big part of your work, right, is advocating for different data, AI solutions.

114

00:15:25,107 --> 00:15:34,455

Brandy: So just in your experience, how does securing the right sponsorship within an organization accelerate the adoption of data-driven initiatives?

115

00:15:36,804 --> 00:15:43,290

Ramila Peiris: I think it helps bringing different stakeholders to the table and gain alignment faster.

116

00:15:43,310 --> 00:15:56,700

Ramila Peiris: If you have a right sponsor with the right message, this will help to bring different stakeholders and bring them to the same goal and gain alignment faster.

117

00:15:57,501 --> 00:16:18,867

Ramila Peiris: there and also the right sponsorship can remove roadblocks right to enable faster execution of such projects like you know if your leader or your sponsor is really really believing in what this all about what your project is and I am sure that you know your sponsor will help you remove roadblocks.

118

00:16:20,133 --> 00:16:28,316

Ramila Peiris: Uh, and also I think, um, I mean, you know, most important part of, uh, you know, uh, data projects is, you know, how do you find some resources?

119

00:16:28,356 --> 00:16:33,217

Ramila Peiris: So, you know, finding the right sponsor can also help us, uh, find resources.

120

00:16:33,997 --> 00:16:41,279

Ramila Peiris: Uh, they can help, uh, they can advocate for and create the right environment to secure user adoption.

121

00:16:42,040 --> 00:16:42,240

Ramila Peiris: Right.

122

00:16:42,260 --> 00:16:49,522

Ramila Peiris: So I think, uh, I think that, you know, if we can cover these aspects, if we, if we have the right, right sponsor sponsorship.

123

00:16:50,461 --> 00:16:51,762

Brandy: Yeah, completely agree.

124

00:16:51,842 --> 00:17:10,493

Brandy: And as listeners are tuning in and they're thinking about how to wrap their arms around early data planning, I'm wondering if you could perhaps share some examples of how early data planning has helped avoid challenges in later stages.

125

00:17:10,513 --> 00:17:13,515

Ramila Peiris: Yeah.

126

00:17:15,290 --> 00:17:29,475

Ramila Peiris: I think for me, if you are building a new manufacturing building, it is so important to consider data as an important part of the building project itself.

127

00:17:30,956 --> 00:17:38,739

Ramila Peiris: So here I'm not just talking about automation systems and other foundational capabilities that capture data.

128

00:17:40,400 --> 00:17:56,065

Ramila Peiris: In addition to that, we must also focus on creating the right capabilities to enable the delivery of clean, organized, ready-to-use data and analytics to the end-users.

129

00:17:56,265 --> 00:18:06,108

Ramila Peiris: From my experience, early planning has helped us to create data capabilities to enable faster problem-solving.

130

00:18:07,029 --> 00:18:26,070

Ramila Peiris: and decision making, which meant we could better adhere to project schedules if it is a new project or new manufacturing facility, or in a tech transfer situation where we are trying to bring a new product into the market.

131

00:18:26,790 --> 00:18:34,575

Ramila Peiris: You know, having the data capability to help us make decision is really, really important.

132

00:18:35,275 --> 00:18:38,337

Ramila Peiris: And having that in a timely manner is even more important.

133

00:18:39,677 --> 00:18:40,758

Brandy: Yeah, absolutely.

134

00:18:40,778 --> 00:18:48,462

Brandy: Incredibly crucial, you know, and just kind of looking, it feels like things are moving at rapid speed.

135

00:18:50,669 --> 00:19:02,146

Brandy: Looking into the future a bit, what advancements in data and AI do you believe will host the greatest impact on the pharmaceutical and manufacturing sectors?

136

00:19:04,751 --> 00:19:11,915

Ramila Peiris: I think any advancement that take us towards performing analytics faster will have great impact.

137

00:19:13,295 --> 00:19:19,558

Ramila Peiris: I mean, more specifically, I think somebody told me this year is the year of AI agents.

138

00:19:20,138 --> 00:19:33,745

Ramila Peiris: So more specifically, I'm looking forward to the advancements in AI agents that can not only do analytics, but also can tackle some of these difficult data engineering problems.

139

00:19:36,585 --> 00:19:39,027

Brandy: So AI agents, that's what you're looking forward to.

140

00:19:39,047 --> 00:19:39,107

Brandy: Yep.

141

00:19:40,388 --> 00:19:40,668

Brandy: Yeah.

142

00:19:42,209 --> 00:19:56,318

Brandy: And just kind of thinking about organizations and where they're at, you talked about kind of starting a manufacturing facility from scratch, but that's a huge financial lift, right?

143

00:19:56,758 --> 00:19:59,700

Brandy: And a lot of these facilities are going to have to

144

00:20:01,428 --> 00:20:05,392

Brandy: kind of go back and retrofit and clean up and do things.

145

00:20:05,432 --> 00:20:11,677

Brandy: I mean, that seems like kind of more of a reality for a lot of folks is like getting themselves prepared.

146

00:20:11,717 --> 00:20:25,809

Brandy: I mean, do you have any advice to those that are in a current manufacturing situation and maybe some initial steps to take to try and to move into a more modernized environment?

147

00:20:26,930 --> 00:20:27,471

Ramila Peiris: Yeah, I mean,

148

00:20:27,978 --> 00:20:35,764

Ramila Peiris: Yeah, I mean, my experience is creating the data capability that I had mentioned during this talk.

149

00:20:36,644 --> 00:20:42,248

Ramila Peiris: And it's not expensive compared to all the assets, everything that you spend on a new building.

150

00:20:44,410 --> 00:20:49,294

Ramila Peiris: Actually, it's very limited budget.

151

00:20:49,334 --> 00:20:55,038

Ramila Peiris: We can create capabilities where the payback time could be

152

00:20:56,938 --> 00:21:00,059

Ramila Peiris: few years, even less than a year sometimes, right?

153

00:21:01,400 --> 00:21:06,141

Ramila Peiris: So I think probably the missing piece is the education.

154

00:21:06,181 --> 00:21:10,482

Ramila Peiris: People don't really think about data as a capability.

155

00:21:10,502 --> 00:21:19,084

Ramila Peiris: When you build a facility, they think about equipment, the assets, the process, how to improve the process and things like that.

156

00:21:19,544 --> 00:21:20,725

Ramila Peiris: They don't really think about

157

00:21:21,685 --> 00:21:27,135

Ramila Peiris: How can I use also the data in this facility to make decision faster?

158

00:21:28,176 --> 00:21:31,943

Ramila Peiris: So I think it's more for education, the thing that we need to do.

159

00:21:33,900 --> 00:21:47,803

Ramila Peiris: Really, creating data pipelines, creating analytics solutions that help system in our new facilities, new buildings, AI ready is not expensive.

160

00:21:48,243 --> 00:21:57,325

Ramila Peiris: I'm talking about it through my own experience, like we build data capabilities in my current role in different sites now.

161

00:21:58,405 --> 00:22:01,186

Ramila Peiris: My team is building like right now two

162

00:22:02,346 --> 00:22:05,028

Ramila Peiris: supporting two brand new manufacturing facilities.

163

00:22:06,049 --> 00:22:11,352

Ramila Peiris: And the budget we have is really, really limited, right?

164

00:22:11,833 --> 00:22:14,254

Ramila Peiris: With a very small budget, we can do a lot.

165

00:22:15,875 --> 00:22:17,076

Brandy: Yeah, that's great.

166

00:22:17,136 --> 00:22:18,277

Brandy: That's impressive.

167

00:22:18,337 --> 00:22:19,638

Brandy: I mean, comparatively, right?

168

00:22:19,658 --> 00:22:26,622

Brandy: The other things that are being invested in these facilities, this is like a very minor budget line.

169

00:22:28,952 --> 00:22:39,218

Brandy: So, you know, as we wrap this conversation Abramilla, I'm wondering if you could just talk through how organizations can position themselves to capitalize on emerging trends.

170

00:22:39,338 --> 00:22:42,460

Brandy: Like, what are some initial steps that they can take?

171

00:22:42,540 --> 00:22:48,324

Brandy: What would you recommend maybe some first tackles like an easy end might be?

172

00:22:51,253 --> 00:22:53,074

Ramila Peiris: Yeah, I mean, there could be different approaches.

173

00:22:53,475 --> 00:22:58,419

Ramila Peiris: I would talk, I would think I would share my approach if I'm doing this, right?

174

00:22:58,459 --> 00:23:02,422

Ramila Peiris: So first, investing in people, right?

175

00:23:02,563 --> 00:23:10,550

Ramila Peiris: Find people who has the domain knowledge, who are capable of systems thinking, like think through a problem.

176

00:23:12,852 --> 00:23:18,138

Ramila Peiris: And, you know, that is really, really a foundational capability.

177

00:23:18,298 --> 00:23:26,767

Ramila Peiris: If you think about, you know, have people, you know, any project, you know, it's, you know, we need to work with people.

178

00:23:26,788 --> 00:23:32,354

Ramila Peiris: So people, you know, different stakeholders, so investing people who are able to

179

00:23:33,975 --> 00:23:48,002

Ramila Peiris: work with others, understand the business problem better, and translate that business problem into technical aspects and data capabilities and how to do that.

180

00:23:49,143 --> 00:23:49,983

Ramila Peiris: So that's one aspect.

181

00:23:50,884 --> 00:24:00,291

Ramila Peiris: And also, I would say, create teams that are passionate about working with end-users and solving end-user problems.

182

00:24:00,331 --> 00:24:09,858

Ramila Peiris: Something missing when we think about all these AI solutions, we don't really think about the adoption pieces of the end-users.

183

00:24:10,879 --> 00:24:11,919

Ramila Peiris: I think it's very important.

184

00:24:13,480 --> 00:24:15,742

Ramila Peiris: And so teams that are really

185

00:24:18,002 --> 00:24:31,800

Ramila Peiris: really work with the end users understand, you know, that mindset that they need to understand the business requirement is really a key factor that will, you know, that will

186

00:24:33,262 --> 00:24:35,283

Ramila Peiris: make any project successful, right?

187

00:24:37,125 --> 00:24:38,846

Ramila Peiris: And then also breaking barriers, right?

188

00:24:39,066 --> 00:24:50,735

Ramila Peiris: Especially leaderships can help breaking barriers and silos to bring different groups from different parts of the organization to work on data and AI related projects.

189

00:24:51,135 --> 00:24:59,942

Ramila Peiris: So oftentimes when we say data and AI related, we think about programmers, people who can code, people who can model.

190

00:25:00,622 --> 00:25:03,804

Ramila Peiris: It's not enough, right?

191

00:25:04,004 --> 00:25:11,367

Ramila Peiris: That is an important part of these projects, but also the people who are actually the experts of business, right?

192

00:25:12,027 --> 00:25:19,671

Ramila Peiris: The domain knowledge, people with domain knowledge are very important for projects like that.

193

00:25:20,511 --> 00:25:36,244

Ramila Peiris: So I think we got to break barriers and silos to bring different parts together to be able to reach our AI goals and to be able to be successful.

194

00:25:38,585 --> 00:25:48,790

Ramila Peiris: And then finally, I would say, create the right strategies for data management, create the right technical strategies for data management and AI that last long.

195

00:25:50,831 --> 00:26:05,678

Ramila Peiris: What I mean by that is, find the right technical stack, find the right technical solutions and stick to them without changing them every year or two, just because

196

00:26:08,779 --> 00:26:19,825

Ramila Peiris: The industry is changing, but you shouldn't be changing your solution every year, which means that you have to redo everything and then it will slow you down.

197

00:26:21,526 --> 00:26:29,130

Brandy: Would you suggest that people seek out a system that is more agile to the changing environments?

198

00:26:29,170 --> 00:26:33,472

Brandy: I mean, it feels like everything is changing so fast that

199

00:26:34,819 --> 00:26:44,785

Ramila Peiris: Yeah, you know, everything will change and so far it is something that is, you know, it's, you know, change is constant.

200

00:26:44,825 --> 00:26:47,887

Ramila Peiris: It's going to change any, you know, I think what I,

201

00:26:48,792 --> 00:26:53,154

Ramila Peiris: would trust in this situation is like the people who can adjust fast.

202

00:26:54,615 --> 00:27:07,542

Ramila Peiris: So that's why I first mentioned when you asked me what are the things they can do to position themselves, the industries, the organization, what are the things that the organization can do.

203

00:27:08,823 --> 00:27:10,284

Ramila Peiris: I mentioned about people because I think

204

00:27:10,987 --> 00:27:23,210

Ramila Peiris: the having the right people with the right mindset is the most important piece because they can tolerate, they can withstand all these changes because they have the right mindset.

205

00:27:23,790 --> 00:27:32,553

Ramila Peiris: So I think, I think people, you know, we, we, we often, you know, we are worried sometimes, you know, AI, you know, we will lose jobs and things like that.

206

00:27:32,613 --> 00:27:33,833

Ramila Peiris: But I think,

207

00:27:35,494 --> 00:27:49,645

Ramila Peiris: Those who are able to translate, make the connection from business to data will be in a more advantageous position in the future with AI.

208

00:27:50,686 --> 00:27:52,548

Ramila Peiris: So invest in people.

209

00:27:52,568 --> 00:27:55,350

Brandy: I completely agree with you.

210

00:27:55,430 --> 00:27:59,513

Brandy: I think that this is 100% a people thing, right?

211

00:27:59,593 --> 00:28:00,334

Brandy: And also just

212

00:28:02,003 --> 00:28:06,847

Brandy: I really loved when you talked about just all the stakeholders involved, right?

213

00:28:06,927 --> 00:28:22,318

Brandy: And like building something that the end users, like everyone is on board with, that you think about all the different folks who are in the process and being able to understand their perspective and coming together.

214

00:28:22,438 --> 00:28:26,181

Brandy: So even if it's not the whole thing as a people,

215

00:28:27,534 --> 00:28:28,695

Brandy: a people issue, right?

216

00:28:28,775 --> 00:28:32,597

Brandy: From the start of like getting the right people and then bringing the right people together.

217

00:28:32,617 --> 00:28:35,318

Brandy: It's great.

218

00:28:35,558 --> 00:28:36,619

Brandy: That's a great takeaway.

219

00:28:36,699 --> 00:28:47,764

Brandy: I think we've been focused a lot on the different technologies that you can bring into your organizations and where to start and data, but kind of bringing it back to people.

220

00:28:50,913 --> 00:28:51,493

Brandy: Thank you, Ramila.

221

00:28:51,533 --> 00:28:53,855

Brandy: This conversation has been excellent.

222

00:28:53,955 --> 00:28:59,197

Brandy: Is there anything else that you want to leave the listeners with before we wrap it up?

223

00:29:03,780 --> 00:29:11,203

Ramila Peiris: I think I started saying that I love the experience where I had the opportunity to solve problems.

224

00:29:13,004 --> 00:29:14,145

Ramila Peiris: And I love that experience.

225

00:29:14,185 --> 00:29:16,786

Ramila Peiris: I think when you think about data and AI,

226

00:29:21,345 --> 00:29:26,988

Ramila Peiris: you know, I think people need to understand it's a, it's a, it's something that will help you.

227

00:29:27,148 --> 00:29:29,350

Ramila Peiris: It's not going to replace you.

228

00:29:30,470 --> 00:29:44,078

Ramila Peiris: And it's, so it's, it's, it's, you know, you owe yourself to, you know, learn that, learn what is capable of and, and be,

229

00:29:45,560 --> 00:29:52,062

Ramila Peiris: be a person that you can make the links on what business need, what AI can do.

230

00:29:52,302 --> 00:29:57,943

Ramila Peiris: You don't need to be an expert in data or in AI to be successful in AI.

231

00:29:59,224 --> 00:30:00,684

Ramila Peiris: And that's the message I can give.

232

00:30:00,704 --> 00:30:02,925

Brandy: That's great.

233

00:30:03,405 --> 00:30:04,085

Brandy: Thank you so much.

234

00:30:04,145 --> 00:30:06,286

Brandy: Thanks for joining Katalyst.

235

00:30:06,406 --> 00:30:09,687

Brandy: This is a wonderful conversation, and we would love to have you back again.

236

00:30:10,867 --> 00:30:11,347

Ramila Peiris: Thanks a lot.