Rising Stars 2026: Kiera Sobolewski 

Kiera Sobolewski, 26 
Specialist, data science & AI 
Canadian Pacific Kansas City 
 
Nominator’s quote: “Kiera has quickly established herself as a leader and innovator within CPKC’s Engineering Technology group. ... Known for her comprehensive knowledge of on-track safety rules and her willingness to challenge norms to uphold high standards, Kiera has earned the respect of peers and supervisors alike as a leader with vision and integrity.” — Kyle Mulligan, CPKC 

Education: Bachelor of Science in mathematics, University of Calgary; Master of Science in statistics, Western University. 
 
Job responsibilities: I develop data science and machine learning solutions for rail integrity non-vital overlay inspection systems, and for the engineering and capital planning teams. A major part of my work involves building models and using data and artificial intelligence (AI) to help make railway operations more efficient and more informed. 
 
Describe your career path. 
While working on my master’s degree, I completed an internship in data analytics in the healthcare industry, which gave me my first opportunity to apply statistical and analytical skills to real-world problems. After graduating, I joined CPKC, where I started in a data analytics role and later transitioned into data science, where I now focus more on machine learning, AI and developing models that can support railway operations. 
 
What was your first job and what did you learn from it? 
Working part time at the grocery store while I was in school. Balancing work and school helped me learn to manage competing priorities and stay organized, which are skills I still use today. It also taught me that even the small tasks matter. Whether I was stocking shelves or helping customers, I learned that people were counting on me to show up and do my part. It was a great early lesson in responsibility, teamwork and the fact that the world runs more smoothly when everyone does their job well. 
 
What sparked your interest in the rail industry? What told you it could be a place for you to thrive? 
I found this job opportunity at CPKC on the company’s career page and it looked like a great opportunity where I could apply what I learned in school to real-world problems. What drew me to rail was the impact. The work is highly practical, and the solutions we develop can have a direct effect on operations, safety and efficiency. The rail industry is a great place data science could be applied to make a real difference, and that made it an exciting place for me to grow. 
 
What’s one of the most valuable lessons you’ve learned in your career? 
Solutions do not need to be perfect on the first attempt. Sometimes, you have to start with something that works, learn from it and continue improving it over time. That mindset has been especially important in data science, where models and systems often evolve and improve as you gain more data, feedback and understanding. I have learned to stay patient with the process, remain open to change and stay motivated to keep looking for ways to make things better. 
 
What are your passions outside of work? 
I enjoy road cycling and have a bunch of long cycling adventures planned for this summer. I also enjoy bouldering both indoors and outdoors.  
 
How do you plan to keep making an impact in your corner of the industry? 
I plan to keep looking for opportunities where data science and AI can help improve efficiency and decision-making in the railway industry. I am also interested in continuing to learn about new AI tools and data science techniques being used in other industries and exploring how those ideas can be adapted to rail. There is a lot of potential for innovation, and I want to be part of bringing practical, useful solutions into the industry. 
 
What’s the biggest challenge confronting the rail industry today? 
One of the biggest is modernizing systems and processes that were not originally designed with AI or advanced analytics in mind. Many legacy systems contain valuable information, but it can be difficult to integrate that data in a way that is clean, consistent and useful for modern data science applications.