CV_GRAM

CV_GRAM is a full-stack CV review, profile storage, and search platform.

It is designed to work as both:

• a helper for reviewing and ranking CVs
• a profile-hosting space where people can turn a CV into a searchable structured presence

The public landing page reflects that directly:

• people can get their CV rated by AI
• people can build a profile similar to LinkedIn
• people can search for others based on CV scores and profile signals

What The App Includes

The current product covers:

• account creation, login, email verification, forgot password, and reset password
• CV upload and AI-assisted scoring
• storing CV-derived profile data
• profile viewing for the current user or another user
• search across saved profiles using text signals and score thresholds
• session management and account security tools

Stack

Frontend:

• Vue 3
• Vue Router
• Vite

Backend:

• FastAPI
• SQLAlchemy
• PostgreSQL
• SMTP email delivery

AI / processing:

• OpenAI-backed CV scoring and profile extraction flows

Project Structure

• frontend/CV_GRAM: Vue application
• backend: FastAPI API, DB models, routers, services
• run_fullstack.py: builds the frontend into the backend and runs the app on one port

Main Backend Areas

• backend/main.py: app entrypoint and frontend static serving
• backend/routers/auth_routers/auth_router.py: auth, sessions, verification, reset password
• backend/routers/v1/parser.py: CV parsing, scoring, and profile generation
• backend/routers/v1/resume_score.py: profile lookup and search

Main Frontend Screens

• / landing page
• /login sign in
• /register create account
• /forgot request password reset
• /reset-password complete password reset
• /verify-email confirm email token
• /dashboard workspace overview
• /score upload and review CVs
• /profile view structured profile data
• /search search saved profiles
• /sessions manage active sessions

Screenshots

Login

The login screen provides the main entry point into the authenticated workspace and links directly to registration and password reset flows.

Register

The registration screen keeps onboarding minimal: email, password, and a short note explaining that email verification is required before protected features can be used.

Password Reset Request

This screen starts the password recovery flow by allowing an existing user to request a reset link by email.

Dashboard / Workspace

The workspace dashboard summarizes the account state, the latest saved CV score snapshot, and the broader workflow supported by the application.

CV Review Upload

The CV review screen supports document upload and exposes the main actions: review a CV, save the score, run an advanced review, and generate a profile.

CV Review Results

After processing, the review screen displays the generated score breakdown together with highlights, strengths, and weaknesses extracted from the uploaded CV.

Profile View

The profile page turns CV-derived data into a structured profile with score summary, identity details, skill signals, experience history, and education.

Search

The search page combines text filters and score thresholds to surface stored profiles and make comparison easier.

Sessions / Security

The sessions screen gives users visibility into active device sessions and lets them revoke individual sessions or sign out everywhere.

Local Development

Frontend only

cd frontend/CV_GRAM
npm install
npm run dev

By default the frontend talks to http://localhost:8000 during dev. You can override that with VITE_API_BASE.

Backend only

Install backend dependencies in your virtual environment, then run:

.\.venv\Scripts\python -m uvicorn backend.main:app --reload --host 127.0.0.1 --port 8000

Full stack on one port

Use the integrated runner:

.\.venv\Scripts\python run_fullstack.py --host 127.0.0.1 --port 8000 --reload

This command:

• builds the Vue frontend
• copies the production bundle into the backend
• serves both the frontend and backend from the same origin and port

Email Links

Verification and password reset emails now derive their public base URL from the incoming backend request. If the app is running on http://localhost:8000, the email links point there automatically. If you run it on another port, the links follow that port.

Database SQL:

This SQL file contains all the necessary commands to initialize it to a new server:

• design/database.sql

Search Test Data

To seed extra users for profile search testing, use:

psql -d your_database_name -f backend/db/seed_search_test_users.sql

The seed file is here:

• backend/db/seed_search_test_users.sql