kangjin2014 / JobAdvisor

An end-to-end project, which accesses the possibility of a resume being considered by HR.

Job Advisor (INACTIVE)

Web scrapping code subjects to the changes of indeed pages. Hence I think the web scraping code no longer work unless you upgrade that accordingly and continuously. On the flip side, I want to address that, the general techs in this project are webscraping as collect data method, elasticsearch as database and matching backend, and flask as a simple front-end to allow file submission. It provides an example of building an end-to-end data related app using purely python.

If you dont have resume handy, please download the resumes in the 'sample_resumes' folder to test this web out.

How to run?

1. Download git and install packages

   $ git clone https://github.com/kangjin2014/job_advisor.git -b devops
   $ pip3 install -r requirements.txt
   

2. Spin up an ElasticSearch database running at port 9200, localhost. Job postings scraped from Indeed will be streamed in. Notice: docker should be installed.

   $ bin/run_elasticsearch.sh
   

3. Kick off job posting download, streaming the data into elasticsearch server.

   $ python3 -m indeed_scraping
   

4. Spin up a Flask web server which accepts the resume from end users

   $ python3 -m flaskr
   
   open the address http://localhost:5000 in the browser
   
   submit the resume by pressing the 'submit' button (support *.doc or *.docx files only)
   
   And will see some recommendation. My suggestion on reference value is based on a small sample study. Will need to impove by machine learning in the future.
   

🐨