Experience
Work Experience
Government of Canada - 3.3 years
- Junior Data Scientist
Canadian Space Agency | 01/2023 - 06/2023, 09/2023 - 04/2024
[Worked as an IT-01]
Was part of the Data and Emerging Technologies team working on various open science initiatives.- Created computer vision algorithms in Python to perform quality analysis of digitization and feature extraction on millions of ionogram films from the Alouette and ISIS satellite missions, for which my team was given the top Government of Canada data award.
- Sped up Alouette data pipeline 9x by implementing Keras Optical Character Recognition (OCR) processing on GPUs using CUDA.
- Spearheaded effort to create agency-wide bilateral MOU tracker dashboard on PowerBI.
- Contributing to both the data strategy working group and options analysis for modern data lakehouse infrastructure solutions for institutional & scientific data repositories.
- Chosen to be an official mentor and judge for the 2023 Space Apps Challenge.
- Data Scientist / Analyst
Global Affairs Canada | 05/2022 - 01/2023 (co-op)
[Started as co-op student and was then promoted to an EC-04]
Made models and visualizations to enhance data-driven decision-making on various files in the Asia Pacific Branch and in support of the Centre for China Policy Research and Coherence.- Used Python to fine-tune GPT-3 and used GPT-3 embeddings for a variety of national security projects including Retrieval Augmented Generation (RAG) to aid with text summarization and grounded question answering using reports from missions abroad.
- Investigated correlation between economic exposure and variables such as their voting trends in the United Nations using Python and PowerBI dashboards.
- Used R for timeseries forecasting of budget use of diplomatic missions abroad.
- Student Researcher
Herzberg Astronomy & Astrophysics Research Centre | 09/2021 - 04/2022
[Part of the NSERC-CREATE New Technologies for Canadian Observatories Training Program]
Applied machine learning techniques in Python, using Keras, to astronomy data processing.- Successfully trained a convolutional neural network to select the best stars for point spread function (PSF) creation, taking only 6% the time of the existing method.
- Begun development of a convolutional neural network to discover new icy minor planets past Neptune using data from the Outer Solar System Origins Survey.
- Defence Data Science Assistant
Defence Research and Development Canada | 01/2021 - 04/2021 (co-op)
Was part of the Center for Operational Research and Analysis.
- Developed novel graph-based algorithms in Python to analyze open source information.
- One of the presenting team members for international hackathon where we won the best presentation award out of 22 teams.
- Renewable Energy Data Research Assistant
CanmetENERGY-Ottawa | 09/2020 - 12/2020 (co-op)
Worked on Remote Community Renewable Energy Analysis and Mapping project to bring renewable energy to Canadian northern and remote communities.
- Automated running existing simulations that utilize historical meteorological data to model potential renewable energy generation and GHG reductions for ~200 off-grid communities in Canada in Python.
- Generated capacity and cost estimates to transition individual communities off diesel, and compiled these results for some of the first Canada-wide estimates.
- Created interactive visualization of results for policymakers using Tableau (learn more and see my first dashboard here).
TRIUMF: Canada’s Particle Accelerator Center - 1.4 years
- Machine Learning Architect Student
Particle Physics Group | 05/2023 - 01/2023, 05/2024 - 09/2024
Joined Antihydrogen Laser PHysics Apparatus (ALPHA), the leading antimatter collaboration at CERN as part of the Azuma Fellowship.
- Designed and trained PointNet-like models in PyTorch to do regression to reconstruct the vertical position of antimatter annihilation events in the ALPHA-g detector based on Monte Carlo simulations, resulting in a model that is able to reconstruct this position to a precision twice as good as the conventional method and near-zero bias (on simulation)
- Participated in shift work for the ALPHA-2 experiment for 2 months which largely involved running physics experiments from the control room and doing real-time data analysis at CERN.
- Attended the CERN Summer Student Lecture Programme.
- Then continued in the fall part-time in the fall to work on a similar project for T2K and having my first introduction to more high performance computing using multi-GPU training on our national compute cluster.
- Photosensor Detector Researcher
Particle Physics Group | 01/2020 - 04/2020 (co-op)
Contributed to detector development for Hyper-Kamiokande, a next-generation neutrino observatory.
- Prototyped a novel neutrino photosensor detector (multi-PMT module) by assembling and testing in the lab, then refining design in SolidWorks, as well as 3D printing and 3D scanning new parts.
- Used Python and C++ to analyze and visualize detector data to detector efficiency.
University of Waterloo - 2.3 years
- Various Roles
University Housing & Athletics | 09/2018 - 12/2024 (w/ breaks)
When attending school in-person, I worked various roles from Residence Ambassador, Climbing Wall Staff, and Residence Don. As a don, I lived in residence at a small college and engaged in various tasks to help make students' university experience as good as possible.
- Was directly responsible for the 19 students on my floor and we won the floor of the year award.
- Patrolled the entire college and responded to emergencies (medical, mental health, etc.) when on duty.
- Provided support to students and resolved conflicts among residents.
- Planned residence-wide events and captained the college intramural team.
MDA Space - 4 months
- Robotics & Space Operations Intern
Guidance, Navigation, and Controls Engineering | 09/2021 - 12/2021 (co-op)
Worked on artificial intelligence algorithms to use on Canadarm2 and potentially Canadarm3.
- Further developed anomaly detection and fault diagnosis programs to run more efficiently using Python and Keras, trained on both MATLAB simulations and real flight data from Canadarm2.
Notes
click to expand
- Most of this work experience was gained through the University of Waterloo co-op program, where we would alternate between school and work each 4 months for throughout our degree and were encouraged to try new things. This led to me doing placements at the Government of Canada, TRIUMF, and MDA.
- You can learn more about my co-op work experiences through an article the University of Waterloo co-op program wrote about me.
- Pretty much all of my technical roles heavily involved Python programming, version control with git (e.g. GitHub, GitLab, and both GUI's and command line tools), as well as working in Linux-like systems.
- I then also worked for my University directly during most of my in-person school terms.
- Also note that so far, most of my non co-op term work has been part-time as I was juggling classes at the same time and for my role at the Canadian Space Agency I took a couple of months off in the summer to pursue research abroad at CERN.
Volunteer Experience
University of Waterloo Data Science Club - 1.6 years
- Host & Editor
Podcast Team | 01/2024 - 09/2024
Listen to our podcast, The Data Den, almost everywhere you find podcasts including Spotify.
- Edited podcasts with guests from OpenAI and Weights & Biases.
- Hosted podcasts with guests from Cohere and Google DeepMind.
- Reading Group Lead
Education Team | 01/2023 - 04/2024
Learned loads about data science by helping to make the topics more accessible to other students.
SEDS-Canada - 2.3 years
- Assistant Project Manager
Canadian Stratospheric Balloon Experiment Design Challenge | 11/2021 - 01/2024
Assisting with logistics for students to fly experiment payloads on high-altitude balloons.
- Provided feedback for proposals and design reviews and answered applicant questions live.
- Organized and hosted live post-flight student presentations.
- Was liasion between student teams and the Canadian Space Agency.
- Led revamping of application process to diversify and increase number of applicant teams to record level.
