Having been CIO at Regeneron Pharmaceuticals for six years, and promoted to that role from within, Bob McCowan had the advantage to approach such an all encompassing position with an intimate knowledge of how the infrastructure runs. “We started on a whole modernization program, so going in as CIO, I was able to pick up on that and move forward,” he says. “And Regeneron is a founder-led organization. They’re still there, and it creates a culture where there’s a bit more risk taking than other organizations I’ve worked for because of the science-based approach, and doing well by doing good. That creates a type of excitement I just didn’t see in some other organizations.”
So with this as a foundation, McCowan has equal parts perspective of archived data, and tools at his disposal to maximize potential value. “This is where the combination of cloud, big data, and bringing it together allows you to look at it all,” he says. “In the past, you may not have seen correlations, but now because that historical data is through ML, it helps influence things we’re doing today.”
As exciting the possibilities are, though, putting everything in check is vital due to increasing hype, but that’s just part of the reality. “Some people take huge programs and almost bet the house on them,” he says. “But others find practical ways that can accelerate and enhance the work we’re doing. It’s good to start there, then build the capability and muscle, and then grow.”
Gen AI is certainly a focal point for McCowan in terms of value, but it takes great imagination as to where that value lies, and the evolving approach needed to puzzle it out over the coming years. “We’re using AI a lot, but our approach to generative is we believe there’s a lot of value in it, yet the value is very different based on the industry. In our industry, for instance, AI isn’t going to replace scientists or the scientific approach. But it will make a good scientist even better.”
McCowan also discussed tech modernization, especially the creativity needed to get the most out of generative AI. Watch the full video below for more insights.
On using data to uplift science: The centralized data platform is the result of a lot of investment over the past few years. But at the simplest level, what we do is research and find drug candidates, and then go to what we call PMPD, or pre-manufacturing, and they learn how to scale it up. They then have to transfer it to manufacturing at large scale. So there’s a whole process from how you take it from research to what we call AIOps industrial operations. We have 12 FDA approved products, so this group knows what they’re doing. But the opportunity arose to go back and look at the process, see how they engage, and what this all does. Essentially, it creates a platform that takes all the data from those processes, and provides it in a way that each person in the step of the process can get access to it. In doing that, we started to see that those groups sometimes spoke a different language. They talked about the same data, but in different ways. The way they transferred data was now legacy based — PowerPoint slides, Excel — or you needed to know who to talk to. By capturing this data in the platform, we simplified that process and made it much smoother. The biggest success was bringing those individuals and subject matter experts together, and it was able to empower them to put this program in place.
On gen AI trends: A lot of the things we’re looking at are practical examples of where generative AI can help. One simple example is, within our industry, you have lab notebooks where you have to track everything. Historically they were done on paper, so we’ve identified some paper lab books with a wealth of information, but it’s not in an easily accessible format. So where we’re using generative AI is to scan and interpret those and read them, and it’s dealing with handwritten notes, pictures, and sketches, but it’s presenting it in a format now that we can expose to many others and capture it. And while the generative AI maybe doesn’t understand at all, it’s able to tag it in a way that others will be able to look at it and see if there’s value.
On cloud to manage data: Not everyone might want to hear this, but it takes a long time to rework infrastructure to transition to cloud. We took a native cloud approach and moved probably 60 to 70% of everything we do to cloud. But we did that very thoughtfully. We identified what made sense to stay on premise. And then in the move to cloud, we also refactored and redesigned it to make sure we took the benefits. So once you get it into the cloud, you suddenly realize you can deal with much bigger data sets, and this idea of connected data comes into play. So the approach we took was we’ve got to have a data platform that’s going to deal with all the ingestion with the quality issues. We’ve got to present the data in the right formats to different groups. Some of them work with research data, others with clinical data or manufacturing with regulatory reform. But it’s about taking a very thoughtful approach to understanding what the process is, how it’s being used, and who’s using it. The other aspect is to be prepared to re-engineer because the technology is moving and processes move so fast. For example, we had an award for high performance computer in the cloud, and since then, we’ve made adjustments to it, but it looks nothing like what it looked like three years ago. So you got to keep revisiting and looking at how people use it and decide if cloud is still the answer, or does it make sense to bring it closer and bring it back in-house. So it’s continually evaluated.
On creating educational partnerships: I personally went out and became OECD certified. Part of it was I wanted to think about what the pressures are there because if you think of what’s happened in the last few years, with Covid-19, for example, the SEC started putting more rigor into things like cybersecurity. So the board had to lean into it. And now you have AI, so just being able to understand how they’re thinking about it helps me shape how I message things. All the boards are slightly different and I mostly interact with the audit committee, and they’re very concerned about cybersecurity. So we do briefings on it all the time. They’re also very interested in AI and we use it a lot. I think sometimes with generative AI, people see that it’s only two years old, but it’s actually been around a long time. But there’s a lot of interest, so we’re doing more presentations and more updates. But as a CIO, you have to put yourself in their shoes and the risks to board members has increased, too. So providing that oversight can be really challenging in the dynamic world that we’re in today.
Read More from This Article: Culture and cloud combine to harness data at Regeneron
Source: News