ResuMix

Job hunting sucks. We make it suck less. ResuMix optimizes and tailors your resume to instantly match your experiences to job descriptions and generate perfect resumes for each role.

Comment

Inspiration

Tailoring resumes for each job application is one of the most time-consuming parts of the job hunt, especially for students and early-career professionals. We realized how repetitive and inefficient this process is, which sparked the idea of ResuMix.

What it does

ResuMix allows users to:

  • Enter all of their experience, skills, projects, and achievements
  • Upload or paste in any job description
  • Instantly generate a polished resume PDF customized for that job

Using a combination of semantic search and generative AI, ResuMix analyzes the job posting and clusters your experiences based on how relevant they are. It selects the most matching points and rewrites them in a way that aligns with the tone, keywords, and responsibilities in the posting, saving time and increasing your chance of standing out.

How we built it

Backend: Python Flask to manage user data and AI processing

AI: OpenAI GPT and Sentence-BERT for semantic matching between job descriptions and user experiences

Frontend: React + Vite for a fast, responsive UI

Storage: NeonDB for user account creation and login

PDF Generation: Used ReportLab’s Platypus library for building dynamic, styled resumes with flowable layouts. Incorporated Canva-designed templates and graphics by exporting visuals from Canva and embedding them as images to enhance resume aesthetics and branding.

Challenges we ran into

  • Matching job descriptions to resume content without just relying on keyword overlap
  • Learning how to implement clustering algorithms to better organize experiences and extract themes across a user’s resume
  • Formatting unstructured AI output into consistent, clean bullet points
  • Getting Platypus to play nicely with variable content while preserving design integrity

Accomplishments that we're proud of

  • Achieved real-time tailoring of resumes based on job postings
  • Designed a clean and intuitive UI that matches industry standards
  • Integrated semantic similarity search + generative rewriting successfully
  • Learned how to apply clustering for smarter AI decision-making
  • Created a smooth user flow from input to instant, tailored resume PDF

What we learned

  • How to leverage generative AI in a practical way for real-world problems
  • The importance of good prompt engineering and filtering AI outputs
  • How to work under pressure and pivot quickly during technical roadblocks

What's next for ResuMix

  • Add LinkedIn scraping to pull experiences automatically
  • Support multiple resume templates/styles
  • Integrate with job boards like Indeed/LinkedIn for 1-click applications
  • Launch a public beta to gather feedback and iterate

Built With

Share this project:

Updates