George James Conidis

Senior Data Scientist at TD Bank Group

Toronto, Ontario, Canada

About

I'm an inquisitive data scientist exploring the reach and abilities of Machine/Deep Learning across various problems. I have over 15 years of experience in Astrophysics, which enabled and encouraged my passion for understanding the physical sciences. I also used this opportunity to allow and help those pursuing higher education, always rising to the occasion to inspire, support, and foster the best in individuals. My aspirations in life are simple; to help make the future one we can all be proud of by making our community and individuals thrive through inspirational actions, meaningful and compassionate connections, and evidence-based discussions on achieving excellence.

Experience

  • Applied Machine Learning Scientist II at TD Insurance
    Sep 2024 - Present · 1 yr 11 mos

  • Data Scientist II at TD
    Sep 2021 - Present · 4 yrs 11 mos

  • Business Development Specialist at Mitacs
    Aug 2020 - Aug 2021 · 1 yr 1 mo

  • Data Science Fellow at Insight Data Science
    Jan 2020 - Aug 2020 · 8 mos

    Insight Data Science Fellows Program is a postdoctoral training fellowship that bridges the gap between academia and a career in data science. Based in Silicon Valley, New York City, Boston, Seattle, Toronto, and Los Angeles, as well as a growing network across North America, the program enables scientists to learn industry-specific skills needed to work in the growing field of big data at leading companies.

  • York University (8 yrs 4 mos)
    • PhD Candidate in Astrophysics
      Sep 2011 - Aug 2018 · 7 yrs

    • Teaching Assistant
      Sep 2010 - Apr 2017 · 6 yrs 8 mos

    • MSc in Astrophsyics
      May 2010 - Oct 2013 · 3 yrs 6 mos

      Created and optimized a search algorithm to identify copies of our host galaxy's (The Milky Way) neighborhood of galaxies (The Local Sheet) in the universe. The results of which are a catalogue of the analogue neighbourhoods (Local Sheet Analogues) along with the new search algorithm. Scientific Motive: Effectively these analogue neighborhoods provide cosmic laboratories to probe questions surrounding our own galaxies formation, evolution and it's effect on neighbouring galaxy companions. Or has the neighbouring galaxies strongly influence the Milky Way's formation/evolution? So was it Nature, Nurture, or a bit of both? Stay tuned for the results of the PhD! Distinct Challenges Addressed: - Identifying the geometric configurations of galaxies despite having low number statistics and non-linear systemic errors associated with the data set (it's not the data's fault, it's the Universe's) - Dealing with data sets which are considerably large (order of millions of rows) - Standardizing multiple data sets to be calibrated on the same system accounting for systemic instrument and post-data reduction biases. Skills Mastered in the Process: - Coding/Scripting/Analysis in Matlab and Python - Effectively Communicating Scientific / Computational Goals and Results to Specialist and Non-Specialist communities - Imaging and Spectroscopic Data Product Calibration, Normalization, and Standardization from different Instruments / Data Repositories. - Statistical Analysis of data products with a special focus on Optimization, Pattern Recognition, Monte Carlo Simulations