Testimonials SEDS alumni

Qixuan Yang (MSc SEDS, 2018)

Qixuan Yang is currently a Ph.D. student in the Department of Political Science at Yale University. Previously, he worked as a senior consultant / senior data scientist, focusing on natural language processing (NLP) and machine learning operations (MLOps).

Why did you opt for the SEDS study program?

The interdisciplinary approach and a wide range of course selections in maths, statistics, computer sciences, and social sciences were very attractive. I liked how the curriculum was crafted, such that the students can grow a deep understanding of the theoretical and practical aspects of data science.

What did you like the most about the program?

I liked the flexibility of the program. It left a lot of room for the students to put together the portfolio they desire. I was able to learn about very basic math and statistics foundations for data science, as well as conduct empirical (applied) research on many occasions. In addition, there is positive peer pressure: I was very happy to study with other brilliant colleagues from whom I could learn a lot.

How did my studies serve my professional career?

The study helped me in many ways. First, a deep understanding of theoretical foundations prepared me to understand newer research that the curriculum couldn’t cover. As the field of data science is evolving very fast, it is difficult for any program to have a curriculum that is always up-to-date. However, having understood the basic concepts, it is easier for me to read and evaluate emergent complicated data science methods.

Secondly, I have had many opportunities to conduct empirical research with real data even in theoretical lectures. I was able to use modern tech stacks to approach a question with acquired methods. It is a challenging program from this standpoint but brought me a high technical maturity above what the market (at that time) would require.

Recommendations for future SEDS

One thing to remember is to combine your interests with the portfolio that you desire. Here, the portfolio may consist of the subfields in machine learning that you are interested in the most, the tech stack that is required for research in such areas (programming languages, software, etc.), as well as the way to communicate your results. Doing one or several seminar papers with real data will help greatly with discovering your interests, as well as growing your soft skills.

And always think beyond the curriculum: As the field of data science is growing fast, your courses might not be able to cover everything. Read papers from renowned conferences, and search for syllabi from other universities of the same course - even during your professional career. This is the only way that you stay tuned with rapid scientific development.

A further comment...

The SEDS program is challenging but flexible: You can get well prepared either for a data science position in the business/non-profit/public sector, as well as for a Ph.D. program that focuses on empirical research. I believe that an interdisciplinary mindset and the ability to „think out of the box“, which you will acquire from this program, will be important at any stage of your career-related to data science.

It is important that you can already get the first insights into the end-to-end data science project: How to set up one, how to collect and store the data, how to process them, how to train and select the model, and how to communicate the results. The ability to pull such an end-to-end pipeline off is what any work related to empirical research would greatly value. I generally recommend you look into docker/containerization techniques and git. I also encourage you to think about building a demo of your favorite data science project.