This application uses Jupyter Notebook, Pandas, NLTK, Scikit Learn, Matplotlib, Plotly, Flask, Bootstrap, PostgreSQL, and a variety of other technologies to help break down my internship search over the past recruiting season (for Summer 2020). I've created a web application to help others track and visualize their applications as well. The site is live for public use at internshiptracker.herokuapp.com.
This project was an opportunity to gain some experience with Scikit and Flask while shedding light on the challenging internship recruiting process. Through my work, I gained a stronger understanding of the effectiveness of different application strategies and gained additional fluency in the libraries listed above.
It contains three main components:
- a Jupyter notebook
- a Sankey diagram
- a Flask app
This notebook contains exploratory analysis and modeling on my internship search process. The main goal here was to use a text version of my emails to create a classifier that predicts whether a given email is career-related or not. Feel free to view it here.
Clone this repo, then navigate to the notebook after running the following commands (ideally using venv):
cd jobSearch
pip install
jupyter notebookThis application allows users to register, input, track, & update all of their internship applications, and will generate a Sankey diagram for them using the most current data about their applications. The app is built with Flask, a variety of Flask plugins, Plotly, and PostgreSQL. It is hosted on Heroku: internshiptracker.herokuapp.com.
Clone this repo, then run the following commands (ideally using venv):
cd jobSearch
pip install
python app.pyVisit localhost:5000 to see the application.
The Sankey flow diagram below represents my Summer 2020 internship search.
- Improve the classifier.
- Allow users to reset their passwords.
- Speed up generation of the diagram by recalculating & storing it whenever a user updates their application.