Alberto Munguia Cisneros came to New York City from Mexico City for the M.S. in Data Science program at Columbia University with a decade of experience in risk management. Here, the 2021 alumnus shares why he decided to go back to school, and how he applies what he learned through the Data Science Institute (DSI) in his new role as vice president of risk analysis and reporting tools in Morgan Stanley’s risk management division.

What brought you to Columbia?

In Mexico, I worked for the National Banking and Securities Commission for 10 years, most recently as general director of risk methodologies and analysis. In my work, I noticed that the banking industries were going through a heavy digitalization process. Topics in [machine learning and artificial intelligence] started being more relevant, and the fintech industry started to gain traction. In order to learn more about these innovations, and take my career to the next level, I decided to pursue a master’s degree with this focus.

My experience was unique in that I came to DSI from working for 10 years. The transition from work back to school can be challenging. But I felt comfortable at DSI, and I found it to be a well rounded program—not just technical, but also applied. I appreciate how students are given the opportunity to use cutting edge tools on real problems.

What were your favorite courses during the M.S. in Data Science program?

I really enjoyed the class in Algorithms for Data Science. It offers a great foundation to start thinking like a data scientist. It provides a framework and “state of mind” for understanding the algorithms that you will use throughout the program. I also found the Exploratory Data Analysis and Visualization class to be very useful. One of the relevant features of data science is to be a good storyteller. This class gives you the tools to do so.  My final semester elective course in Applied Deep Learning also stands out. It introduced me to the latest tools in TensorFlow, and let me learn from someone who is developing these tools and using them on a daily basis.

Tell us about your capstone project.
The capstone project was actually awesome to do during the pandemic—it offered me a way to continue interacting with colleagues and classmates. Our team worked with management professor Jorge Guzman, building on his work using text-based machine learning to measure startup strategy and performance. We worked with data from over 12,000 startups, comparing early statements with behavior and outcomes over time. In the process, we learned a lot about text-based data analysis methods, and how different algorithms and models affect results. We found that more research is needed. Reality is messy, and it was great to work on a real problem, with real data. 

How does data science influence your work in risk management?

Both risk management and data science are focused on data. Being a risk manager is about identifying risk, measuring risk, and making decisions based on the risk. It is like being a data scientist in that you are working with data and extracting valuable information from it.  

As a risk manager, you see a lot of information. Sometimes it’s difficult to interpret the signals that are in front of you. You need to make a business decision that will impact the operation of your organization or financial institution. All the data science methodologies—modeling types of variables,  identifying outliers or anomalies in the data—are important and useful to the real world of risk management.

There is no secret sauce. The foundation is to spend quality time with your data. Know your data, and build a model from it. Sometimes you need to do a lot of data cleaning, wrangling, and engineering first. The cornerstone of data science is to get to know your data.

— Karina Alexanyan, Ph.D.