As a team of engineering students, we know the stress of having to balance the full course load of engineering during a school term with applying to jobs for co-op. In order to get the best chance of getting an interview, a tailored resume is essential to maximize your chances. However, with so many class deadlines to handle, the goal of creating tailored resumes is often left unmet.
Through an internal survey with EngHack participants, we collected information regarding how much time students spend working on resumes during a school term. We found that:
- 76.9% spend over 1 hour per day on co-op
- 84.6% apply to over 75 jobs per term
- 46.2% spending 30+ minutes on a resume per posting
- 53.9% create 40+ unique resumes over the course of a co-op term
This inspired us to create the ResuMate service that can streamline the process to create tailored resumes for co-op postings by selecting the most relevant summary of experience points. We focused on this section most hiring managers will often look at an application for roughly 5-10 seconds. In this time, the summary of experience and the list of skills are often looked at to get a high-level overview of the candidate.
Overall, with the power of our proprietary phrase parsing algorithm, we are able to identify the best summary of experience points for any given job posting. Our goal is to do the job, to get you the job.
What it does
This platform will take in a job posting PDF through the GUI, ingest the data for parsing, ingest the stored list of summary of experience points, removes filler words from both arrays, identifies keywords, correlate posting’s keywords to the existing summary of experience keywords, runs proprietary phrase comparison algorithm to compare variants of each word for similarities, generates similarity scores for the summary of experience points, ranks summary of experiences, ingests how many summary of experience points that you would like on your resume through the GUI, and displays top 3 points onto the GUI that can be copied into the users resume.
How we built it
We built this platform using Python, Figma as well as Python libraries including NLTK, Levenshtein, PyPDF2, PyQt5, and docx. We used the PyCharm and Visual Studio Code interpreters to develop the front end and back end, respectively.
Challenges we ran into
As our team is separated across a 3-hour time zone difference, we found it challenging to continually be updating others of current progress to optimize the integration of developed code sections such as our front end and back end.
Accomplishments that we're proud of
Accomplishments that we are proud of include our ability to work together in our respective areas of expertise, despite the distance and 3-hour time zone difference between our team members. Throughout this hackathon, we continually pushed ourselves to develop new levels of functionality throughout our platform, including the ability for our keywords to include synonym correlation, which improves the accuracy of the system by detecting variants of the same skill.
What we learned
From this experience, we learned how to work with PDF files and how to extract information for data parsing. This aided us in the endeavour of reading existing job postings and correlating the information to existing summary points from the user.
We also learned how to work with several new Python libraries such as Levenshtein and NLTK. These libraries were essential to include the desired functionality of our MVP deliverable.
Additionally, this was the first time our team worked with Python GUI packages to create a functional interface. This provides a simulated experience for the user to see what interactions they would have with our web portal in a fully deployed service.
What's next for ResuMate
For ResuMate, our next steps include:
- Create web portal, integrate ML models and deploy pilot projects with users
- Refine ML modes for engineering-focused areas
- Build customer base & branch out to other job areas
- Create strategic partnerships with other resume and job finding services