Inspiration
We noticed that a lot of fresh graduates who do not have job hunting experience are not aware of what kind of resume the hiring managers wish to see, especially in their field. This causes some of them not to showcase themselves efficiently in the resume even though they are actually a competent candidate. We hope to have a platform to close the gap between fresh graduates and industry hiring expectations.
What it does
Synopsis can scan through your resume with a neural network that is able to analyze if your resume is fit to the company's expectations based on your field. It suggests your top 3 best suited companies with your qualification level indicated on a scale from 1 to 10. It also shows you where your resume is placed compared to the people in the same field in a distribution curve.
How we built it
The website is built using HTML and CSS. We collect a set of dummy resume that we annotate them as an employee or not employee of the top 3 companies. After that, we train our neural network using Google Machine Learning and store the model in Google Storage. Our users will upload their resume to the Cloud through our website. The Cloud Function will find and scrape the important information from the resume, the machine learning algorithm will provide a score for the resume uploaded. After obtaining the scores, the Cloud Function will write the score to the proxy in Firebase real time database created by the website once the resume is read. Any changes on the proxy will be updated immediately on our website.
Challenges we ran into
Lack of training data and lack of man power
Accomplishments that we're proud of
Tailored designed resume scoring algorithm for applicants' interested companies, where most of our competitors provide a resume scoring based on overall scoring system.
What we learned
Data collection is one of the most crucial part of machine learning and is the most tedious task.
What's next for Synopsis
Synopsis can streamline the application process by providing a virtual "resume guru", giving you suggestions to upsell yourself. To further improve our grading system, we will need aid from job hunting sites like Indeed and hiring companies to provide us with data to help improve applicants' resume quality. We recon that resume building is a hassle and everything is in LinkedIn, we also aim to automatically design and update your resume from your LinkedIn profile.
Log in or sign up for Devpost to join the conversation.