Resumo

Automated AI tailoring and integrated job tracking for students.

Comment

Resumo

Inspiration

The modern job hunt is a full-time job. We noticed that the biggest bottleneck isn't finding roles—it's the soul-crushing task of manually tailoring your resume for every single application to beat the ATS (Applicant Tracking System). We built Resumo to give job seekers that time back, turning a manual hours-long process into a few seconds of AI magic.

What it does

Resumo is an end-to-end career copilot.

  • Smart Parsing: Upload your "master" resume once. Our engine uses Gemini's native PDF capabilities to parse it into a structured, searchable profile.
  • Job Tracking: Manage your entire pipeline with a Kanban-style dashboard. Drag jobs from "Saved" to "Applied" to "Interviewing."
  • AI Tailoring: With one click, generate a hyper-optimized resume tailored to a specific job description. It enforces a "one-page rule," selecting the most relevant experience and rewriting bullets for maximum impact.
  • ATS Score: Get instant feedback. Our analyzer scores your tailored resume against the JD, highlighting matched keywords, missing skills, and actionable gaps.
  • Live Diff Viewer: See exactly what the AI changed. Review edits side-by-side with your base resume before finalizing.
  • Professional PDFs: Export download-ready, ATS-friendly PDFs directly in your browser.

How we built it

  • Frontend: React 19 with Vite and Tailwind CSS 4.0 for a high-performance, modern UI.
  • Backend: Supabase handles Authentication and the PostgreSQL database.
  • AI Engine: Google Gemini (via @google/genai) powers our resume parsing, tailoring, and ATS scoring.
  • State Management: Zustand for lightweight, reactive editor state.
  • PDF Generation: @react-pdf/renderer for client-side document creation.
  • Drag & Drop: @dnd-kit/core for the interactive job board.

Challenges we ran into

PDFs are where data goes to die. We initially struggled with manual text extraction from complex resume layouts. After experimenting with traditional parsers, we moved to Gemini's native PDF processing, which significantly improved accuracy for non-standard layouts. Fine-tuning the "one-page" constraint was another hurdle—ensuring the AI prioritizes the right experience while keeping the output concise required extensive prompt engineering and strict JSON schema enforcement.

Accomplishments that we're proud of

We’re really stoked about the Live Diff Viewer. We didn't want the AI to be a "black box," so we built a feature that highlights exactly what the AI changed compared to your base resume. Combined with our ATS Match Score, it gives users total confidence that their resume is truly optimized before they hit "Download."

What we learned

The secret to reliable AI applications is structured output. Moving from simple text prompts to enforcing strict JSON schemas (using Zod for validation) turned a "neat demo" into a predictable, robust tool. We also learned that providing users with "human-in-the-loop" editing capabilities is essential—AI handles the heavy lifting, but the user always has the final say.

What's next for Resumo

  • One-click Cover Letters: Automatically generate matching cover letters for every tailored resume.
  • Browser Extension: Save job postings directly from LinkedIn, Indeed, and Greenhouse with one click.
  • Interview Prep: AI-generated mock interview questions based on your tailored resume and the job description.
  • Template Library: More professional PDF designs to choose from.

Built With

  • lucide
  • react
  • supabase
  • tailwind
Share this project:

Updates