JobMatching — AI-Powered Job Platform

A full-stack job platform with AI-powered resume matching, real-time job search, and employer tools. Built with FastAPI, React, PostgreSQL, Pinecone, and OpenAI.

---

Features

• Job Board — Browse 500+ real jobs scraped from the Adzuna API across 10 tech categories
• AI Resume Matching — Upload your resume and get ranked job matches powered by OpenAI embeddings and Pinecone vector search
• Smart Filters — Filter by location, job type, remote, and salary range
• Sort Options — Most recent, Highest salary, Best AI match, Most relevant
• Job Modal — Click any job card for full details, requirements, and a direct "Apply now" link
• Employer Dashboard — Post and manage job listings
• Seeker Dashboard — Upload resume, view detected skills, see AI-scored matches
• Dark / Light Mode — Persistent theme toggle in the navbar
• Auth — JWT-based login and registration with role-based access (seeker / employer)

---

Tech Stack

Backend

Layer	Technology
Framework	FastAPI (async)
Database	PostgreSQL 16 + SQLAlchemy (async)
Migrations	Alembic
Auth	JWT (python-jose) + bcrypt
AI Embeddings	OpenAI text-embedding-3-small
Vector Search	Pinecone
LLM	OpenAI GPT-3.5-turbo via LangChain
Job Data	Adzuna Jobs API
Cache	Redis
HTTP Client	httpx

Frontend

Layer	Technology
Framework	React 18 + TypeScript
Build Tool	Vite
Styling	Tailwind CSS + shadcn/ui
Animations	Framer Motion
State	Zustand (auth + theme)
Data Fetching	TanStack Query (React Query)
Routing	React Router v6
Forms	React Hook Form + Zod

Infrastructure

Service	Technology
Database	Docker (postgres:16)
Cache	Docker (redis:7-alpine)

---

Project Structure

job-platform/
├── backend/
│   ├── app/
│   │   ├── core/          # Config, database, dependencies
│   │   ├── models/        # SQLAlchemy models (User, Job, Application)
│   │   ├── routers/       # API routes (auth, jobs, users, matching)
│   │   ├── schemas/       # Pydantic request/response schemas
│   │   ├── services/      # Business logic (embeddings, AI agent, scraper)
│   │   └── main.py        # FastAPI app entry point
│   └── alembic/           # Database migrations
├── frontend/
│   └── src/
│       ├── components/    # Reusable UI components
│       ├── pages/         # Route-level page components
│       ├── stores/        # Zustand state (auth, theme)
│       ├── types/         # Shared TypeScript interfaces
│       └── lib/           # Utilities (api client, helpers)
└── docker-compose.yml

---

Getting Started

Prerequisites

• Python 3.11+
• Node.js 18+ (or Bun)
• Docker Desktop
• Accounts for: OpenAI, Pinecone, Adzuna

---

1. Clone the repository

git clone https://github.com/your-username/job-platform.git
cd job-platform

---

2. Start infrastructure (PostgreSQL + Redis)

docker-compose up -d

This starts:

• PostgreSQL on localhost:5433
• Redis on localhost:6379

---

3. Backend setup

cd backend
python -m venv .venv

# Windows
.venv\Scripts\activate
# macOS / Linux
source .venv/bin/activate

pip install -r requirements.txt

Create backend/.env:

DATABASE_URL=postgresql+asyncpg://postgres:postgres123@localhost:5433/jobplatform
REDIS_URL=redis://localhost:6379
SECRET_KEY=your-secret-key-here

# OpenAI — https://platform.openai.com/api-keys
OPENAI_API_KEY=sk-...

# Pinecone — https://app.pinecone.io
PINECONE_API_KEY=pcsk_...
PINECONE_INDEX_NAME=job-embeddings

# Adzuna — https://developer.adzuna.com
ADZUNA_APP_ID=your-app-id
ADZUNA_APP_KEY=your-app-key

Run the server:

uvicorn app.main:app --reload --port 8000

API docs available at http://localhost:8000/docs

---

4. Frontend setup

cd frontend
npm install    # or: bun install
npm run dev    # or: bun dev

App runs at http://localhost:5173

---

5. Seed jobs

With the backend running, call the seed endpoint to populate jobs from Adzuna across 10 tech categories:

curl -X POST "http://localhost:8000/api/jobs/seed"

This fetches ~500 real jobs (software engineer, frontend, backend, data engineer, ML engineer, DevOps, cloud, mobile, product manager, data scientist).

---

6. Embed jobs for AI matching (optional)

To enable the "Best match" AI sort, embed all jobs into Pinecone:

# Requires a logged-in user token
curl -X POST "http://localhost:8000/api/match/embed-jobs" \
  -H "Authorization: Bearer YOUR_TOKEN"

---

API Endpoints

Method	Endpoint	Description
POST	/api/auth/register	Register a new user
POST	/api/auth/login	Login, returns JWT
GET	/api/jobs	List jobs (search, filter, sort, paginate)
GET	/api/jobs/{id}	Get a single job
POST	/api/jobs	Create a job (employer only)
POST	/api/jobs/seed	Seed jobs from Adzuna
GET	/api/users/me	Get current user profile
POST	/api/users/me/resume	Upload resume text
POST	/api/match/from-profile	AI match jobs to saved resume
POST	/api/match/embed-jobs	Embed all jobs into Pinecone
GET	/health	Health check

---

Environment Variables

Variable	Required	Description
DATABASE_URL	Yes	PostgreSQL async connection string
REDIS_URL	Yes	Redis connection string
SECRET_KEY	Yes	JWT signing secret (generate with openssl rand -hex 32)
OPENAI_API_KEY	Yes	For embeddings and LLM matching
PINECONE_API_KEY	Yes	For vector search
PINECONE_INDEX_NAME	Yes	Name of your Pinecone index
ADZUNA_APP_ID	Yes	For job scraping
ADZUNA_APP_KEY	Yes	For job scraping

---

Quick Demo

https://youtu.be/Xo9_CMKXPZE

---

License

MIT