Matthias Le Dall

I use a combination of machine learning and quantum computing tools to forecast market trends and client behavior that assist traders across Capital Markets, M&A, Fixed Income, and Treasury optimize investment strategies. I use database management systems to query private communication data, macro indicators, forex, equities, bonds and company fundamentals. I collaborate with stakeholders to understand business needs, I design live dashboards to deliver insights, and I work with engineers to automate data pipelines.

I am developing Stackify, a recommendation product that suggests semantically related posts by identifying key concepts on a StackExchange posting. I am using natural language processing (NLP) tools, such as scikit-learn tf-idf, Google's word embedding Word2Vec, and Gensim TextRank for keyword extraction. In addition, I use a combination of clustering and dimensional reduction algorithms, such as k-means and PCA, for classification and visualization.

I developed the CAN-YES project, a Canada-wide online-based astrophysics outreach program whose objective is to connect students from grades 4 to 12 across Canada to astrophysicists working in observatories around the world. Thanks to my work, our organization allowed 100 students across Canada to connect remotely with scientists at the CFHT, the JCMT telescopes in Hawaii, as well as with the HAA institute in Victoria, BC.

I study the atomic structure of magnetic impurities in proximity to superconductors, in order to understand the noise spectrum in superconducting quantum interference devices (SQUIDs). SQUIDs are used as a possible implementation of quantum bits used for quantum computers, thus a better understanding of the noise allows to improve the stability and reliability of quantum computers.

I developed models for particle interactions beyond the standard model of particle physics, focusing on the phenomenology of light new particles. I used those models to explain the matter-antimatter asymmetry in the universe by describing the evolution of the abundance of particles via a set of Boltzmann equations. This research conferred me strong skills in quantum field theory as well as scientific programming in both Python and Mathematica.

When magnetic impurities are in proximity to superconductors, the proximity effect leads to low-energy states, called Shiba states. Using a combination of numerical, analytical and effective field methods, I show that Shiba states have a complex energy spectrum inherited from the orbital structure of the underlying impurity. The complex spin and orbital character of the Shiba states implies a rich response to electric and magnetic stimulation, thus implying extensive sources of noise. This work can be applied to identify sources of decoherence in superconducting devices, particularly relevant to quantum computers that use superconducting qubits. Through this research, I perfected skills in quantum many-body physics, as well as scientific programming in Mathematica and Python.

I studied the impact of diffusion processes on the efficiency of directional solidification in crystal growth.

I was in charge of running Taekwondo training sessions to students of all ages, two to three times a week. This experience gave me strong communication and leadership skills. The intensive and extensive practice of martial arts taught me the values of perseverance and hard work, which are now part of my work ethics.