Societal transformations are often epoch-defining milestones in mankind’s history. The transition from Stone Age to Bronze Age. The changeover from the Middle Ages to the Renaissance. The overthrow of Communism in favor of Free Market economies & democracy. The fall of the Iron Curtain was a particularly defining event for this episode’s guest, who grew up in Poland & witnessed his nation’s conversion to capitalism after the collapse of the Soviet Union. What he experienced then not only continues inspiring him, but also influences the advice he gives organizations on digital transformation.
As the person in charge of Data Analytics & Artificial Intelligence for one of the world’s biggest IT and business consulting services firms, Tomasz Jamroz is uniquely positioned to be an enterprise change agent. He draws upon his vast expertise to share with us some fascinating use cases he’s worked on, the 2 key factors organizations must address before undertaking digital transformation, and the technologies most likely to advance our capabilities & impact our world over the next 10 years.
Guy Nadivi: Welcome everyone. My name is Guy Nadivi and I’m the host of Intelligent Automation Radio. Our guest on today’s episode is Tomasz Jamroz, VP Consulting and Head of Data Analytics and Artificial Intelligence at CGI, one of the largest IT and business consulting services firms in the world. Tomasz is a thought leader in emerging technology strategy and execution, and he has a specialized focus in advanced analytics, artificial intelligence, the internet of things, and big data platforms.
Guy Nadivi: With a resume like that and being based out of Canada, we wanted to get Tomasz on the show to share some insights with us on what he sees happening in the world of digital transformation from his advantage point, north of the border. Tomasz, welcome to Intelligence Automation Radio.
Tomasz Jamroz: Guy, thank you for having me. I’m delighted to be here.
Guy Nadivi: Tomasz, please tell us a little bit about yourself and how you ended up in a senior position involved with AI at one of the world’s biggest consulting firms.
Tomasz Jamroz: Sure. Maybe where I should start is … And that will help the audience understand the funny accent they hear. I’m Eastern European. I actually grew up in Poland, and I had a chance to go through a significant transformation very early as a teenager when Poland’s move from being a communist country to become a free market and democracy republic. Experiencing that transformation so early created that I had a special relation with change. I always see a very positive aspect of any change that is occurring around me and this is something that really helped me to move and progress in my career.
Tomasz Jamroz: By education, I’m Neurobehavioral Psychologist. One could say that’s not a typical AI education or a background that you would expect for somebody in my position, but for me, Neurobehavioral Psychology was a way to discover, and learn, and build methodology that helped me progress. I think it’s very important when you look at the education as not really what you’re learning, but you are able to derive from the journey. I really like Neurobehavioral Psychology, but I haven’t had 100% conviction towards the discipline.
Tomasz Jamroz: I shared that with my profs when I was writing my undergraduate thesis and they advised me that, “Tomasz, you have this very logical approach and we’re currently building this interdisciplinary team with computer science department where your knowledge about cognitive psychology and neuroscience would be a perfect pair.” This was for me the first time to become a part of such a diversified team, and learn how I can utilize something that I have learned in a particular domain and apply it to something totally different.
Tomasz Jamroz: I got actually pretty excited about computer science, to the extent that I convinced the dean of computer science to enroll me in the fourth year computer science courses without having proper prerequisites and background. I almost failed, but it was a very valuable lesson. Very quickly I realized that Neurobehavioral Psychology is not really what I want to be doing. What I want to be doing is, I want to be helping organizations and I want to help organizations transform or improve them.
Tomasz Jamroz: I became the System Architect. I architected custom solutions for HR and IT departments, but I realized as I was doing it that I was still missing something. I realized that what I’m missing was the knowledge about business and I decided to do MBA. Throughout my MBA, I was absolutely fascinated by the concept of ERP. Looking at how you can take any organization and you can transform it and fit it into the ERP or SAP or any other, and run it successfully was something that really seemed very interesting and a great opportunity for me to go into and learn more. I became SAP consultant.
Tomasz Jamroz: Very early in my SAP consultant stage of my life, I discovered a niche and the niche was something that brought me closer to, we can say, automation. I realized that regardless organization you work with, there is multiple departments that translate into multiple different modules within ERP. However, there is one denominator that is the same. The denominator was that any of those processes, when it fails, it creates an exception and that exception prevented from a system automatically completing process from A to Z.
Tomasz Jamroz: I decided, could I collect all those exceptions? Could I find any anomalies in those exceptions, and identify that the exceptions are exceptions only because the system thinks that they are. But in reality, if we slightly modified a few parameters, they’re not exceptions anymore. As such, I was able to create a very interesting niche. Utilizing a lot of data input, I was able to help organizations decrease the volume of exceptions by 20 to 30%. When you think about it, all those exceptions, when they’re creating the system, at the end they’re sent to the back office where there’s number of people that they need to resolve them.
Tomasz Jamroz: The fact that you were able applying those simple analytical methods to resolve such a significant portion, was something that really made me fascinated about unlimited opportunities that there are when you really utilize the data and you find the trends and patterns to improve. At the same time, I was actually … At this stage of my career, I was part of Accenture Consulting. Accenture decided at the time, it was around 2014, pivots to the new digital economy. There was actually a great book written by Omar Abbas who was Chief Strategy Officer, and architect behind that transformation.
Tomasz Jamroz: Anybody who is interested of how to move and transform organization, you’ll find a lot of interesting information there. As Accenture decided to pivot, I felt the need and maybe that’s because it goes back to my appraisal that I’ve seen transformation, I’ve seen that there’s good things coming out of transformation. I decided to do the same, and I discovered a human-centric design. I spent a lot of time learning and getting involved in automated testing in robotic process automation, in advanced analytics, but it wasn’t until I joined my first data science team.
Tomasz Jamroz: I remember working my first data science team, and we were building this segmentation for the customer, looking at their clients. I engage in that discussion with my team and I started asking questions about centroids and distance between those centroids when we identify different clusters. I remember the team looking at me and asking, “Tomasz, you told us you don’t know anything about machine learning.” And I said, “That’s true, I don’t. But for me, this is statistics.”
Tomasz Jamroz: It’s a very funny story because I have this vivid memory when I was actually starting and I remember the advanced statistical methods course. Taking that class and studying K-means, which is what you usually use for clustering and segmentation, I was wondering “Why am I doing it and when and where will I ever need it?” Little did I know, this was something that helped me establish respect with my data science team, and then my career really progressed very quickly.
Tomasz Jamroz: It was pretty early stage of digitalization. I had a very good background and it enabled me to jump on those opportunities and I got engaged in multiple very interesting initiatives in the big data platform space in cognitive aspect. I was working on more and more complex projects and I really liked it.
Tomasz Jamroz: By total surprise, I was contacted and asked to come and interview by CGI. I was happy at the place I was, at the time. Out of respect of CGI, it’s a great organization, I decided to take that interview. I remember I felt that that interview didn’t go well, because when I came to the interview, I’d been proposed a particular role and I politely declined. I shared what I would like to do, or what I see the challenges that exist within the market are, and how I could help any organization resolving those. I remember leaving that interview, I was like, “Well, good that I have a job because I don’t think I’m going to have any other interviews anytime soon.”
Tomasz Jamroz: But to my surprise, I’d been contacted back and I had a chance to meet another person. We dived deeper and discussed about what is my vision and I found it very interesting. I found it quite fascinating that I am now having a discussion even though I rejected what was offered to me, and the organization did not take it defensively. Quite opposite, they decided they would like to understand better the idea that I had. We had a number of discussions that actually led to and ended with a job offer that was put in front of me.
Tomasz Jamroz: I remember when that happened, I found myself puzzled because I didn’t … I wanted to do it, but I was afraid and it was the moment when I realized that I actually became comfortable with what I was doing so far and this opportunity is a new challenge. The moment I realized, I had to do the same one I’m always advising to my clients that stagnation kills innovation. If you have an opportunity that scares you, it’s something that usually is good for you and I did. I’m very happy that I did. CGI is a big multinational organization, but at the same time its core is an organization that allows a lot of entrepreneurship.
Tomasz Jamroz: Probably it’s the fact that our founder started the organization from a team of two to what it is today, almost 80,000 consultants and it kind of happened over a short period of time. I’m very happy because it allowed me to join the organization that I can address the challenges that are there, and I can have a full support to do it. Pretty much, that’s my story.
Guy Nadivi: It really sounds like your whole life has led to a career focused on transformation. So, I’m curious Tomasz, if you can talk about some of the more interesting use cases you’ve been involved with at CGI and the quantified impact that digital transformation had on the organization’s you worked with.
Tomasz Jamroz: Absolutely. I’d like to share three different examples. One example I would like to share is coming from the manufacturing industry, which I think has a great potential for digitalization. Another is actually applying a particular type of technology across different industries. My last example would be from the utilities, but it’s interesting example because example that comes and it’s something that was fully built by one of our clients. I was really fascinated when I learned about it.
Tomasz Jamroz: Let me start with the first example. Everybody sees digitalization and when we look at the back and front office, there is a no brainer that that’s where the great opportunities for automation are. I would encourage everybody to look in any industry that is asset heavy, because that’s where especially advanced analytics can be very helpful. One of the examples that I would like to share with you today is Uddeholms. Uddeholms is a producer of high alloyed tool steel. They’re based in Sweden, and they came to CGI with a very particular challenge. They wanted to decrease number of cracks in the steel, and they identified that the steel cracks are really affecting their profitability.
Tomasz Jamroz: When we engaged with them, it was very interesting because very quickly we realized that they actually have a lot of data points that they collected around their manufacturing process. What they were missing was a proper platform that would allow ingestion and analysis of the information that they already had in different forms. When we started working with them, we first concentrated on establishing that foundation. Utilizing IoT devices and collecting all the information, including the historical data that they had, to build this platform.
Tomasz Jamroz: Once we collected all the information, we started to simulate and run different models, and develop algorithms that allow us to identify exactly when and where the cracks occur. Once we were able to simulate it, we were able to identify where within the process are the key elements, the gene to modify look for or change, in order to prevent those cracks. At the outcome of that model, we were able to predict with 70% accuracy when the cracks will occur, and as such we were able to decrease the waste that otherwise would be unprevented.
Tomasz Jamroz: This is an example that we often see in manufacturing industry, in pharmaceuticals. This is something where you can really utilize the vast amount of data that a lot of organizations is collecting today, and use it to build predictive analytics, digital twins, and so on.
Tomasz Jamroz: Another example, and I find it’s very interesting, is the concept of a machine vision. Machine vision in real time analytics, it’s something that really allows us to augment and enhance a lot of areas that we are already operating in. I’m going to give you an example of retail. We work with one of the consumer electronic giants, and they had a very successful targeted marketing campaign, and it shouldn’t be a surprise. Marketing using some advanced analytics for a long period of time, but what we decided to do is, could we augment the customer experience within the store with real time analytics and in particular with machine vision.
Tomasz Jamroz: What we very quickly realized that, when we were combining information that coming with POS, and we were enhancing it with the information of how many customers actually interacted with the new product that is on the shelf, what was their sentiment and what was the nature of their interaction? We were able to enhance and increase the sales using this targeted advertising now enhancing with machine vision by 120%. The concept of machine vision, it’s something that is really taking off. We work with a number of cities to develop smart cities.
Tomasz Jamroz: Cities receive a lot of CCTV feeds. There is a lot of monitoring in a lot of cities around the globe, and machine vision is something that can help increase the safety of the citizens. We develop the models that we were able to identify what are the accidents and where are they occurring, and what is the best way for emergency crews to be dispatched to it. We used the machine vision in another industry in renewable energy.
Tomasz Jamroz: One of the challenges with the wind turbines is that the maintenance is a critical aspect. Normally you have to dispatch your workforce to go and check regularly the equipment, and those are big constructions and there is safety hazards and so on. We’re able to utilize drones and machine vision to collect the information to identify where really the problems are, and as opposed to dispatching people just to go and check the structures and the wind turbines, we were able to dispatch them exactly to where the problem is and advising them what they need to take in order to resolve that because we’re able to provide some insights about the nature of the challenge. This is something that we see quite a bit.
Tomasz Jamroz: The last example I would like to bring is actually coming from one our clients. Hydro-Quebec, it’s one of biggest renewable energy power generators in the world. They’re based in Montreal, and they have their own research institute. That research institute has developed a fault location algorithm and using advance data analytics, they were able to detect and diagnose both permanent and momentary faults. Those faults are hard to identify. Those are network issues, and those network issues are actually the root cause behind many outages and ultimately customer complaints.
Tomasz Jamroz: What we engaged with Hyrdo-Quebec is, we actually realized it’s such a great solution that we would like to help them to commercialize it. I think this is really what speaks volume about the state of the industry. I think it’s really emerging domain. You normally are not surprised to see companies like CGI, a technology power house that is able to come and build AI-driven solution. But I’m very happy to see that our clients, on their own, are capable and they’re reaching a certain level of maturity to build solutions utilizing advanced algorithmic capabilities themselves.
Guy Nadivi: Tomasz, listening to you talk about all those different use cases, I’m curious. It’s been said that the failure rate for digital transformation projects can be as high as 84%. What is one thing you think organizations undergoing digital transformation must do to ensure that they’re a part of the 16% that succeed?
Tomasz Jamroz: Of course. This is actually a very interesting question and this is something that at CGI we … We actually perform our own study and what we do is, we have something called the voice of our clients. Through the voice of our clients, we interview the executives around the globe. It’s over 1,500 executives from different industries, different countries, where we’re trying to understand what are the challenges. One of the questions that we ask is, what are the challenges that you experienced and identified as your top challenges in the area of digital transformation.
Tomasz Jamroz: We actually see that 83% of people we interview, they share with us that culture change and change management is a key aspect of a transformation. I think this is something that is very important. Addressing the people in the digital transformation should be a key. I often see that when we talk about the digital transformation, there is a lot of emphasis that is put on the word “digital” where in reality we should really concentrate on the transformation.
Tomasz Jamroz: How do we transform the people? How do we transform the process? How do we transform the organization? Making something digital is fairly simple. You can take a piece of paper with text, put it on the scanner, and you have a digital format of the sheet. Making people to adopt and open themselves to embrace transformation is much more significant and much more challenging.
Tomasz Jamroz: The second thing, and this is like 70% based on our research is that, technology and especially legacy technology. A lot of organizations still today, they’re dealing with leftovers of mainframe computing and all the other custom solutions that allow them to grow and allow them to be where they are today. However, the lifespan for those technologies is really getting to the abrupt end, and a lot of organizations realize that when they really want to go and fully embrace the digital, they have challenges, their technology is not capable of taking them.
Tomasz Jamroz: On the positive note, when you look at those two major factor which are culture and the legacy, what we see asking the same questions and starting that year after year that you see an improvement. More organizations realize that the culture change and change management is a key in the digital transformation, as well as more and more organization address the challenges of having legacy systems. I think this is something that is incredibly important for any organization to address if they really want to be successful.
Guy Nadivi: Listening to you discuss the many different technologies and industries that you work with at CGI, I’d love to hear what you think are going to be some of the biggest disruptions that we’ll see in three, five, or even 10 years from now with respect to automation, AI, and other digitally transforming technologies.
Tomasz Jamroz: Okay, a million dollar question. I wish I had the answer that I can be 100% certain and invest in stock accordingly. But let me try to address it looking at what happened up to now and how do I foresee the change appearing in the nearest future. When I look at the first three years, I think that the greatest adoption or the biggest change will come from the adoption of the technologies already available to us. I feel that the maturity of a technology that is surrounding us is greater than our adoption. When you think about it, the concept of a PC that move the computer from a research center universities close to us with mobile, with augmented, with wearables, we really have ability to utilize the technology in very close proximity to where we are.
Tomasz Jamroz: We also created a lot of data which I don’t fully see that we’re utilizing today. When you think about it, we have creating over 40 zettabytes of data which actually translates into, we created more bytes of information than there is stars in observable universe. We have the vast majority of information available to us. We have the tools how we can access it. The limitations that I still see that we have today, and I don’t think they’re going to be resolved in the next three years. I think it’s more five or maybe in the full decade is our computing power.
Tomasz Jamroz: We’ve made amazing progress. However, we’re reaching the point that I foresee that the biggest change will come from some kind of a new computing power that is introduced. It could be quantum computing. The concept of quantum bits, they would exponentially increase our computing capability. One of the great examples when you think about the power of quantum computing is that encryption packages that we use today, if we were to attack them with a brute force, it will take us 200 years to crack. If we applied quantum computing to the same challenge, it would take us a month.
Tomasz Jamroz: I think that illustrates what is the odd of a possible that can happen. I think that is what really is required to answer something, what we call general AI. To enable artificial intelligence to have this heuristic capability of inheriting the information and becoming a more general or more true intelligence.
Tomasz Jamroz: The last thing that I think is very interesting is … I’m also curious how that field progresses is the edge computing or maybe even distributed edge computing. I think that at the moment we’re going to get a moment in time when we’re going to get to the point that we need to make a critical life or death decisions and milliseconds and really play the role. I think there’s a potential of the edge computing to really take off, and become a significant force, having an impact on the transformation and of future of technologies.
Guy Nadivi: Interesting. Tomasz, for the CIOs, CTOs, and other IT executives listening in, what is the one big must have piece of advice you’d like them to take away from our discussion with regards to engaging in digital transformation?
Tomasz Jamroz: My biggest word of advice is that any IT executive today should think like they are CEO of a technology company. This is something that I believe it’s critical, and the digital transformation is something that really defines be or not to be for any organization. This is something that when I look & I often discuss with different executives is … My biggest advice is always to be honest with yourself in terms of what is your maturity. At the same time, you need to understand what is the art of the possible.
Tomasz Jamroz: What you want to make sure is that in no point in time you allow that gap to really grow. Meaning that you have a certain level of technological agility, but by no means you match what is the potential of that technology. Because what this big gap is causing is, it’s actually decreasing the barriers to entry for your competition and for other technologies, technology companies to go and enter your business. This is something that for me is critical, that IT executives help drive IT organizations and they don’t look at it through the old paradigm that IT is just a cost center.
Tomasz Jamroz: IT should be a differentiator and the value creator for any organization, and IT executives are actually in the most important elements of that success. I’m not jealous of the challenge they face with. They play a very significant role and it’s very important that they become a part of any business strategy discussion and they also help drive the business.
Guy Nadivi: You raise a good point in that, I think it’s sometimes lost in the shuffle, that with all these digital transformations going on, it’s often IT that itself is undergoing the most radical, digital transformation of anyone. All right, looks like that’s all the time we have for, on this episode of Intelligent Automation Radio. Tomasz, thank you so much for coming in today and sharing your thoughts with us. By the way, I’m a big fan of Canada and especially your national dish poutine.
Tomasz Jamroz: Awesome. You should come to Montreal and we can continue the discussion and grab some good poutine.
Guy Nadivi: Sounds good to me. Tomasz Jamroz, Vice President of Consulting and Head of Data Analytics and Artificial Intelligence at CGI. Thank you for listening everyone and remember, don’t hesitate, automate.