yitingchien.web [at] gmail [dot] com | Based in Erlangen, Germany

Casey: A Chrome Extension Applying Machine Learning to Help New Graduates with Job Search

AREA: Chrome Extension & Web Design

TIME & TEAM: A 3-week project with Xinbei Hu

CONTRIBUTION: Involving in the whole process from research, ideation, iteration to implantation with another interaction designer

 

Challenge and behaviour of online job-searching

CHALLENGE

For new graduates who just finish their studies, they rely greatly on online job information rather than word-of-mouth recommendations, since they are still new to industry and lack networking resources. However, job information on the internet is quite a big amount of data and to browse them is an endless journey. According to the interviews with target users and desk research, people want to know how their skills match job requirements. It takes time and effort to search for interested openings and go through those job descriptions. How might we assist new graduates to find the job opportunity that suits them from a huge amount of data? In addition to this, how might we support job-hunter to manage and track those information

How the extension works with a website containing job description

CONCEPT

Casey is a chrome extension that determines how well users match to any job description based on their résumés. It helps people quickly make a decision and learn what they can do to increase the likelihood that they'll get respond by the tips. By operating as an extension, it allows users to compare information between multiple webpages, which is considered as a common behaviour in terms of online job searching. The chrome extension is supported by Natural Language Processing technology to analyse descriptive and interpretative content. The algorithm helps to calculate the overall matched skills versus missing skills to get the final match rate percentage. At the same time, users receive structured information including hard-skill, soft-skill and experience to understand the context of match rate.

ON-BOARDING PROCESS

1. Sign up through email address or Linkedin account 

2. Upload resume as fundamental reference

3. Weight attributes for more accurate matching scores

4. Success page and instruction for the next steps

A system map shows all features that support user journey from on-boarding, scanning job pages to job list management.

Job list management page allows users to record and track interested openings.

Details page of a saved job for further information 

Wireframes iteration round 1, from on-boarding to scanning

scan

track

archive

Wireframes iteration round 2, from scanning to data management

Building the interactive prototype to test with users with premade content to mimic the functionality of machine learning 

Using InVision to build the prototype: https://invis.io/PXBZDAVUW