SKILLMATCH.AI: NOVA SBE workproject

This repo cotains the source code for Skillmatch.ai, a web application that operationalizes the skills-first approach in the labour market. It integrates functionalities such as resume screening, tailored interview questions, bespoke apology letters, and role similarity analysis, offering a comprehensive solution for skills-based hiring and career development. Built with a scalable architecture, the application addresses recruitment inefficiencies and skill gaps. Although limitations such as scalability constraints, reliance on external APIs, and the need for real-world data remain, future iterations aim to refine infrastructure, enhance usability, and expand applicability, advancing the adoption of skills-first practices across industries.

Instructions:

There are 2 options for running the web application locally on you computer. You could 1) fork or clone the Github repo, activate the conda environment and run it directly with flask (this approach might have some dependency issues depending on your OS) or 2) (recommended) download the latest docker image of the application from dockerhub and run the docker container.

1. Method 1: Clone repo and run with flask

Setting up the conda venv

Create the environment with the config-file

conda env create -f environment.yml

Conda should now have set you up with all the necessary dependencies to run the project. Please activate the environment:

conda activate workproject

You could also enable a jupyter-kernel if you want to use this environment in jupyter notebooks:

conda python -m ipykernel install --user --name workproject --display-name "workproject"

Run the we application locally

To run the flask application locally on your machine (first navigate to the user_interface directory)

flask run

2. Method 2: Run the Docker container

Download and run the docker image

1. If not already installed, install Docker for your OS
2. Log in to docker in terminal

docker login

1. Pull this docker image (might need sudo first)

docker pull victorbjorsvik/skillmatch_ai:latest

1. Run the container (might need sudo first)

docker run - 5000:5000 victorbjorsvik/skillmatch_ai:latest

or if you want to add the environment variable for using the admin user:

docker run -5000:5000 -e OPENAI_API_KEY="<your openai api key>" victorbjorsvik/skillmatch_ai:latest

These steps should sucsessfully host an instance of the webapp which you can interact with through your browser.

Technical Overview

• Backend: Flask for handling routes and application logic.
• Frontend: Jinja2 templating for dynamic web content
• ML Models:
  • SentenceTransformer (all-mpnet-base-v2) for embeddings and cosine similarity.
  • spaCy for skill extraction with custom EntityRuler patterns.
• Database: SQLite for managing job postings, applicants, and analysis results.
• APIs: OpenAI API for generating tailored interview questions, coding exercises, and apology letters.
• Containerization: Docker for seamless deployment.

Overview of the application

File tree:

Home page:

External Recruiting (before analysis):

External Recruiting (after analysis):

Tailored Questions (before analysis):

Tailored Questions (after analysis):

Bespoke Apologies (before analysis):

Bespoke Apologies (after analysis):

Similar Roles (before analysis):

Similar Roles (after analyis):

Collaborators

• Hanna Borchgrevink Pedersen
• Tim Gunkel
• Irene Abbatelli
• Luca Oeztekin

Acknowledgements

Big thanks to Amira for letting us use her code as a building block for our project: