Retraining as a Data Scientist: Gaëlle Sellin’s story

Future graduate of ENSAI’s Master for Smart Data Science class of 2024, Gaëlle Sellin has already lived several professional lives. After working as a logistics manager, she became an engineer at CNRS, and is now putting her data science skills to work for the Odyssey team. This project, led by five academic institutions including Inria and Ifremer, aims to better model ocean dynamics.   

Here’s a short highlight of her experience after a semester of intensive courses and a few weeks into a six-month internship.

The journey to data science

Gaëlle Sellin: “Before joining ENSAI’s Master for Smart Data Science, my first career path was as a logistics and transport manager, which led me to carry out several projects abroad. I developed an interest in data science around 2015, since forecasting is critical in the logistics business. Also, new tools had appeared on the market.

On top of that, I had kept the taste for statistics and probabilities developed during my earlier studies. I always thought that if I were to change careers one day, it would be in this field. And that’s what happened 4 years ago! After working in logistics, I joined the Laboratoire d’Océanographie Physique et Spatiale in Brest, in the Ocean and Climate department, as a CNRS engineer.

As part of my career change in data science, I attended courses that, despite their quality, didn’t cover the fundamentals properly, particularly in mathematics. The field of data science offers many learning opportunities. This democratization of knowledge is a good gateway… but it can also lead to pitfalls! Knowing how to compute an f1 score, for example, isn’t enough if you don’t know what’s behind it. So in all honestly, you can be completely mistaken and create your own biases!

The Master for Smart Data Science, a demanding program

I knew that the Master for Smart Data Science was highly regarded as a demanding program, reflecting the image of ENSAI. I was convinced that it would provide me with the fundamentals I needed.

I can confirm that its reputation is well earned! Everything is designed to ensure the quality of the learning experience: the size of the class, the diversity of the students’ backgrounds, which complement each other, and above all the quality of the teaching. It’s a great opportunity to attend classes given by teaching and  staff,research who are also accessible and keen/dedicated/committed to passing on their knowledge.

We know that in 5 years’ time, the world of AI and data science will have changed, and the tools and parties involved will no longer be the same. This course teaches you to be independent in your learning, to ask questions and to seek out information, for example, by reading scientific articles. Mastering the basics makes you autonomous. I think I’m now equipped to adapt, which wasn’t necessarily the case before.

I can already say that I wasn’t disappointed by any of the courses, the program is of a high standard/is top-notch. I really enjoyed François Portier‘s Machine Learning course, because it’s an essential foundation, as well as Valentin Patilea‘s Functional Data Analysis course. This last course was hard for me, as I didn’t have the basics to start with, but I’m continuing to work on it. It really sparked my curiosity, and I was able to develop a critical approach to data.

An internship with the Odyssey team

I’m currently doing an internship at Inria, working on the Odyssey project (“Ocean DYnamicS obSErvation analYsis”), which involves modeling the dynamics of the physical ocean.

I’m working on generative AI, particularly on probability models, flows and diffusion models, which are currently quite popular.  The idea is to use these methods to bring stochasticity to the generation of data from existing deterministic models.

The ocean as well as climate fields are vast and offer really interesting projects. Intellectually, it’s very stimulating, and AI and data science are tools that have a genuine role to play and will continue to develop.

My professional aspirations for the future? Following my own path where I currently am suits me well, but that won’t prevent me from developing personal projects, particularly in the field of sound!”

More about the Master for Smart Data Science at ENSAI