Experience
Computer Engineering and Computer Science Co-op · MITRE
At MITRE I worked in the underwater department. My primary project was for NOAA (the National Oceanic and Atmospheric Administration); they needed to identify what conservation efforts for North Atlantic Right Whales to invest in. I worked on the Aeiral subteam (tactic where whales are counted in the ocean from an aircraft) in a SCRUM workflow, where I simulated the efficiency of aerial surveys.
Highlights
- Served as primary software engineer for a Python-based whale detection model, synthesizing complex climate, aerial survey, and behavioral datasets (NetCDF, JSON, GRIB).
- Directed technical outreach with SMEs in oceanography, data science, and marine biology to procure validated data and refine model architecture, translating research papers into a reliable model.
- Leveraged High-Performance Computing (HPC) to execute large-scale bootstrapping and sensitivity analysis, quantifying the impact of every environmental covariate on the model.
What I worked on
I developed a Python whale detection model. The model takes heterogeneous oceanographic/weather datasets and past aerial survey/behavioral records. The script generated a map of the east coast, with detected and non-detected whales flagged under varying environmental conditions. To make the results trustworthy I ran large-scale bootstrapping and sensitivity analyses to ultimately go on HPC clusters, isolating the contribution of every covariate.
What I learned
The most valuable lesson was how to talk to and learn from domain experts - oceanographers, data scientists, marine biologists - and translate their findings, datasets, and expertise into the model I am developing. This project had about 20-30 people involved, with my subteam being about 3-5 people. I also learned how to give bi-weekly SCRUM stand-ups to the leaders and bi-weekly more complete updates at the whole team meetings.
Photos
