Find the best talent with ScreenDoor

Leverage cutting-edge Natural Language Processing algorithims to recruit the best talent to government.

Get Started

Benefits

ScreenDoor wants to solve problems facing federal government hiring managers.

Save Time

85% of government hiring managers say the hiring process
is a burden. ScreenDoor cuts down the time it takes to evaluate applications.

Made for Government

ScreenDoor is tailored specifically to the needs and practices of Canadian federal government managers. Upload application PDFs and receive consistent scored results.

Easy to Use

Convenient web-based application can be accessed from any computer using a secure login.

Open and Transparent

ScreenDoor is free and open-source software to ensure transparency and support deployment across government.

Unique Features

ScreenDoor's features are tailored to the needs of federal government hiring, based on interviews with managers and aimed to reduce repetition and unnecessary paperwork.

Automated Grading

Applications are graded out of 5, based on experience and asset experience criteria, including evaluating information from written responses.

Information Extraction

ScreenDoor's Natural Language Processing technology extracts semantic information on the depth and recency of an applicant's experience.

Applicant Sorting

Easily access specific job postings and lists of applicants, sort by name and rank, and access summaries and analysis of submitted applications.

Architecture

What's behind the ScreenDoor.

Front-End

Back-End

Database

How It Works

ScreenDoor's intuitive interface runs on Django, the Python-based web framework trusted by Nasa, Instagram, Mozilla and others to deliver fast, secure, and stable web applications. Django allows our Python Natural Language Processing algorithms to interface directly with the brains of our web application, hosted on a Canadian Amazon Web Services instance.

Applying Natural
Language Processing

ScreenDoor utilizes the Natural Language Toolkit (NLTK), an open-source Python library offering named entity extraction, context identification, and other key features. It won't just save you time by tabulating yes/no answers to identify whether applicants qualify for the position; thanks to advances in NLP technology, ScreenDoor will also help you assess an applicant's experience and assets. After processing, you will have a list of candidates, an evaluation of the depth and recency of their experience, and summaries of their written responses.

Prototype

See our design in action!

Project Timeline

Where we are going

Month 1

Project proposal, research, and web application set-up. Establishment of data models and database design. Testing NLP libraries.

Month 2

Development of core ScreenDoor features, such as parsing and processing processing job information and submitted application forms.

Month 3

Refinement of core features and development of complex features such as written response evaluation and scoring applications. Implementing views in front-end and connecting front-end interface with back-end processing.

Month 4

Winding down development, applying finishing touches or any additional features permitted by time. Demonstrating final product to clients, and launching the application for testing in government.

Meet Our Team

The people behind ScreenDoor.

Jared Ridyard

Carleton University, Computer Science

Naman Sethi

Carleton University, Software Engineering

Sam Heaton

Algonquin College, Computer Programmer

Contact Us

If you have any question, don't hesitate!

This template is made with by Colorlib | Some icons from icons8 | Website created and hosted on GitHub