Built a personal data tracking system that records daily soda consumption via an automated startup script and stores entries in a database, with a Jupyter notebook used for exploratory analysis and visualization of intake trends.

Project code is here.
Jupyter Notebook can be viewed here.
Developed an image classification model in Python using TensorFlow to predict the country of origin from Google Street View images, achieving ~20% accuracy across multiple classes.
Project code is here.