workbringer

A resume optimizer balancing the convenience of AI and the originality of user-generated contents. It is geared toward tech job applicants with an overflowing resume, mindful of data leak to AI.

Comment

Inspiration

Inspiration came from applying to 50 jobs on WaterlooWorks. Tedious. Considered trying resume generators out there, but all of them use AI to optimize your resume and rewrite it into shit. So, here I am.

What it does

Like the name suggests, workbringer brings opportunity for work, with speed, originality, and peace of mind with personal information. Like most resume optimizers out there, it uses AI to gradually speed up the process. However, the AI does not see nor edit any user-generated content.

How we built it

The heart of the project is the backend. To keep the resume part of the project clean of AI, I decided to use a split approach. Each item in the resume would have two parts: content and attribute. The attributes are summarized and provided to AI for parsing along with the job description. After that, a dynamic programming algorithm selects the items that will appear on the final program.

Challenges we ran into

In short, flu and overestimation. I planned to do the project with a group of 3, and here I am, soloing the project. As a result, plans had to be cut short. As the person who came up with the idea in the first place, I had a ton in mind. Realizing I am alone, I still chose to do too much stuff. These decision costed me a lot in terms of time wasted researching things too hard to do, and more.

Accomplishments that we're proud of

I am proud of the level of completeness to the project. It has a fully functional backend, and a mostly working frontend to showcase it. I am also very proud of the project idea. The more I worked on it, the more I realized that it is something I need in real life very much.

What we learned

Flask, React, Typescript, and a bucket of Pythons (let's prey they wont bite) Learned the necessity to split code to small segments and to very clearly define datatypes and i/o of functions. Gained more experience putting a website together with a custom API. Learned enough ReportLab to be a solid Latex believer (minus the DevOps part please) Learned to never give up, however shit things are, it will get more shit but you will get more ready.

What's next for workbringer

The project got more momentum than a storm, it will go on. Our goal is either a web service or a ios app. Sounds fun

Built With

Share this project:

Updates