Data Scientist
This phase consists in formatting, cleaning and sampling the data. To gain insight, spot local patterns and explore data structure, I use visualization techniques. Based on the problem that needs solving, I create a baseline model with the right algorithm, that will provide a starting point.
The second phase of a project consists in transforming the attributes or creating new ones, using a mix of linear and non-linear transformations, decomposition, aggregations and domain knowledge. I use a testing framework to evaluate the features and algorithms, and select the ones that are most useful for our model.
In this phase, I focus on hyper-parameters tuning for the best performing models, test the individual performance and the performance of an ensemble of algorithms. After that, is time to launch the model in production by adapting it to an operational setting.
My favorite thing is to find a good story, from a book, a movie, a video game or from data. My passion is to find the Whys and the Hows in economics, business and technology. Working with data provides insight into patterns and correlations that answer those questions. My desire to discover the whole story from the data lead me to getting certified in Machine Learning.
When I feel I've spent too much time in front of the computer, I go biking, hiking or snowboarding. But the thing I love the most when it comes to recharging my batteries, is to travel the world with my wife.
A selection of my work
Feel free to contact me if you want to discuss your project. Check out machine learning competitions I participated in, my repository or my profile below.
Designed by Kevin Powell
