Practicum Spotlight
A candid conversation with five current MSDS students discussing their internship experiences at fintech companies including BlackRock, First Republic Bank, Freedom Financial Network and Opportun
Background and Introduction
Ronica: Hi All! I am excited to talk to everyone today! Can everyone share a little bit about their background and what company they are doing their practicum at?
Isabella: My name is Isabella. I graduated from UC Berkeley in 2020. I did a double major in philosophy and economics. I’m currently working as an intern at Black Rock. Our project deals with time series analysis and we’re trying to build a forecast model to predict the ETL flow of the company.
Ronica: Wow, that’s cool!
Surbhi: Hi, I’m Surbhi. I’m from India. I’ve worked for about 5 years in the data science industry. And now, I’m interning with Freedom Financial Network. My work mostly revolves around building financial products. We are trying to replace the work that the company is currently transferring to another analytics company. So, it mostly involves working on machine learning models and optimizing different types of optimization algorithms available.
Ronica: Oh, that’s nice! Monty, do you want to go next?
Monty: I am currently working at Opportun. It’s pretty similar to a small bank. The company also issues loans and credit cards to customers. So they are a feedback company that cares a lot about their customers. Users can have multiple accounts, so one of our projects is around customer account matching. You can imagine that a user may have multiple accounts. We want to match them as the same user so that we can reduce the risk of sanctioning multiple loans to the same person and better understand user behavior.
Ronica: Great! Wei do you want to go next?
Wei: Before coming to USF, I was a business analyst and financial analyst in China for a security company. I also got my master’s degree in statistics there. For my practicum, I work with Monty at Opportun. Apart from what she mentioned, we are also working on another project to analyze credit card transactions.
Ronica: That sounds great!
Hi, I’m Ronica. I did my undergrad in engineering and I worked as an Analyst before coming into the program. For my practicum, I am currently working at First Republic Bank (FRB)! We are working with a very interesting use case at FRB. Our problem is an NLP-based problem wherein we are working to extract relevant information from bond prospectus PDFs.
Choosing the Practicum
Ronica: Why were you interested in this company as your practicum company? What was it that you felt during Pitch Week that helped you make your selection?
Surbhi: I wanted to work in the finance domain. I have limited knowledge of this area, and I was quite interested in understanding how things work in this area. Other than that, I liked the presentation of our practicum mentor, Jim. During my conversations with him, I realized that he had a great knowledge of the domain and data science, and I could learn a lot from him. So I would say, for me, it was a combination of both the company and the mentor.
Ronica: Yeah, that’s nice! I feel that having a great mentor is very important, especially as we’re starting as data scientists.
Wei: I chose Opportun because I had some experience in the financial industry. Before coming to MSDS, I had no idea about what data science and machine learning was. So I felt that, compared with other candidates, choosing a financial company would give me some comfort as I would not have to worry much about the domain, and I could focus on the data science aspect.
Ronica: Oh wow! That is a really smart way to make the choice.
Isabella: I chose Black Rock for quite a few reasons. First, because it has a big name. I wanted to work for a relatively large institution, to observe how the modern financial business operates. The second reason is the project that I am working on is a brand new project designed specifically for the interns. I think it’s a very good opportunity to be the project owner and carry it from the start till the end. The third reason is the team that we’re currently working with is called the ‘Financial Technology Transformation Initiative’. Basically, our team solves financial problems with technology. We come up with several technology initiatives, such as transforming all the financial data from local to warehouses, building warehouse design, etc. I think it’s a very interesting team to work with. And, lastly, the project. Building a forecast model is a very good experience to apply what we’re learning in class to the industry.
Ronica: Oh, yeah, I agree, I think finance still has a lot of room for adapting technology. And I think it’s a good space for us as data scientists to come into.
I am amazed to see how BlackRock fits all your requirements.
Ronica: For FRB, the first thing that interested me was — hands down — the pitch. Two people from the team presented in person, which was great. Plus, their presentation on how they aim to use technology and state-of-the-art algorithms for managing financial documents was really exciting. We also have two alumni on the team, so that was great to know. Apart from that, like you mentioned Isabella, having ownership over our projects was something really exciting about First Republic as well.
Tools and Technologies
Ronica: What other technologies that you all are using in your internship? Is it something that we use in the program?
We all might have had to learn some technologies that the company specifically uses, but how easy or difficult was it to transfer learning from what we’re doing in school?
Surbhi: I’m working mostly with Python. At school, we are working on Github but in the office, I’m working on Bit Bucket. We’re also working on Google’s BigQuery to fix some data. We used Google Collab in Yannet’s class, which was helpful to get started on Google AI. We can use the technology that we are learning in our courses directly at our internship, so I’m very thankful for for that.
Ronica: Oh, that’s nice. With Python libraries, is there a specific library that you use at work that we didn’t work on in class, or do you use the ones that we use daily?
Surbhi: Most of them are the same. The only thing that was new to me was some specific optimization libraries in Python.
Ronica: That is great to hear!
Wei: The two main technologies that we use are SQL and PySpark. An interesting aspect is the pace of our projects — it matched our class schedule at school. After I learned SQL in Diane’s class, our first internship project was regarding using SQL to extract some data from the database. After that, we learned Pyspark with Diane, and we used it with EMR in our second project.
Monty: That is true! Apart from that, we use Python regularly, especially for exploratory data analysis and visualizations.
Ronica: It’s good that it’s been moving side by side. For me as well, I am using SQL and Python daily, something we use at school too. Apart from that, I use a lot of image-based and text-based libraries because we have image data and we just want to do text extraction. Yannet’s advanced machine learning classes covered a lot of the things that we were working on, so that was helpful.
Isabella: We’re mainly using Python to do the modeling part of my job. For some tasks at the beginning, I even borrowed some code from our homework. So that’s good. I think our classes covered the majority of the technology we use in our practicum daily.
Practicum Experience
Ronica: How has your overall internship been experience so far?
Wei: When we selected our top choices in practicum companies, I knew nothing about machine learning except what we were taught in school. I was interested in creating models to make some predictions. But as Monty also mentioned about our projects, the projects are not just ML. They include data engineering, processing and a lot more technical hands-on work which I did not expect. It has been a great learning experience.
Ronica: For sure! I think this is the most important aspect of working in industry — to solve a lot of problems before getting to the actual problem itself.
Monty: Yeah! After working on the credit card project, I feel I am also now open to looking for different roles apart from a data scientist too, perhaps as a data engineer.
Ronica: Great! Now you have more options than a lot of other people.
Monty: Also, I feel that our mentors are very helpful and very supportive. I like that. Before this, I had no experience in the fintech industry. So the project also helps us understand how data science works in fintech. This can give me some examples to share during my interview processes.
Ronica: It’s nice that you have supportive mentors!
Surbhi: In my experience, I would say it’s a mixed bag. I got to learn a few things which were very new to me. I used to work with optimization problems, but I learned and tried out a lot of statistical optimization methods in my internship. It was good for me to revisit all the probability and statistics. On the other side, I was expecting a little bit more machine learning like everyone else. But, I think that many people have this experience in any company. In terms of my mentor and the data science team, we can approach them whenever we need to, so that is great. They are always available and help us in resolving things.
Ronica: Yeah, that’s true! I feel like when we were doing data science projects in class, we were always given clean data and our first step would be modeling. The practicum gives us a lot of insight into what our job as data scientists would really be like.
Ronica: I had a similar experience at FRB. The first project that we had was just string data and we had to make sense of that. That was something we did not know how to tackle at first. Building a model took around two weeks but preprocessing that data took about three months. Overall, I feel we’ve had an amazing experience! Both with the work that we are doing and the experience of working with such a great team. We have had ownership of the project, and they let us experiment with whatever we feel could work. They’re always available and super-ready to help us, and it’s been a great journey so far.
Isabella: My experience so far has been pretty pleasant. We have a great project and a great team. Our mentor is not a data scientist. He’s actually a software engineer, so when we were building this forecast model, sometimes he was not able to give us too much mentorship in terms of the machine learning aspect, but he has always encouraged us to reach out to other people in and outside of our team within the company. So yeah, it has been great so far!
Ronica: That is great! I am glad that everyone is having a great experience at their practicum companies. We have about three more months of our internships to go, and I hope they are a great learning experience for all of us. Thank you for sharing your experiences and such exciting projects with everyone!
