Resume Tailor Backend

AI-powered backend service for tailoring resumes to specific job descriptions, optimizing for ATS (Applicant Tracking Systems), and exporting polished documents.

Deployed on Google Cloud Run.

---

✨ Features

• Resume tailoring using Google Gemini API
• Job description analysis (URL or pasted text)
• ATS compatibility checks (before & after)
• Job fit scoring (keywords, coverage, gaps)
• DOCX export of tailored resumes
• Stateless, production-ready API

---

🧠 Architecture

• Framework: FastAPI
• AI Model: Google Gemini
• Deployment: Google Cloud Run
• Frontend: Next.js (Vercel)
• Language: Python 3.12+

---

📁 Project Structure

resume-tailor-backend/
├── main.py              # FastAPI entry point
├── services/            # Resume parsing, scoring, ATS logic
├── requirements.txt     # Python dependencies
├── Dockerfile           # Cloud Run container build
└── README.md

---

🚀 API Endpoints

POST /upload

Upload a .docx resume.

POST /tailor

Tailor the resume to a job description.

Input

{
  "resume_id": "string",
  "jd_url": "optional",
  "jd_text": "optional",
  "company_url": "optional"
}

Output

{
  "tailored_text": "...",
  "change_log": [],
  "suggestions": [],
  "ats_before": {},
  "ats_after": {},
  "fit_score": {}
}

---

POST /export

Export the tailored resume as a .docx.

---

🔐 Environment Variables

Configured in Cloud Run, not committed to GitHub.

GEMINI_API_KEY=your_api_key_here

---

🛠 Local Development

python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --reload

---

☁️ Deployment

Deployed using Google Cloud Run via container builds.

This repository is used for source control only — deployments are handled through Google Cloud.

---

📌 Notes

• Generated files (exports, temp data) are excluded from version control
• Authentication and persistence are handled by the frontend
• Designed to be stateless and horizontally scalable