Auto Job Applier

Auto Job Applier is a LinkedIn job-search and application assistant with a powerful Python automation bot, a scalable FastAPI backend, and a modern React admin dashboard built for production SaaS environments.

The backend controls the bot, stores configuration, manages resumes, streams live logs, and exposes dashboard APIs. The frontend is a Vite/React dashboard for editing search rules, uploading resumes, viewing applied jobs, and monitoring the bot in real-time.

---

🏗 Architecture

The project has been refactored into a scalable production SaaS structure with clear separation of concerns.

Project Structure

.
├── runAiBot.py              # Main LinkedIn automation bot
├── server/                  # FastAPI backend (layered architecture)
│   ├── api/routes/          # API route definitions
│   ├── auth/                # Authentication logic
│   ├── bot/                 # Bot controller and process management
│   ├── config/              # Configuration management
│   ├── database/            # MongoDB client and migrations
│   ├── middleware/          # CORS and request middleware
│   ├── models/              # Data models and schemas
│   ├── repositories/        # Database access layer
│   ├── resumes/             # Resume file registry
│   ├── services/            # Business logic layer
│   └── utils/               # Shared utilities
├── frontend/                # React + Vite dashboard
│   ├── src/api/             # API client
│   ├── src/components/      # React components (auth, dashboard, jobs, etc.)
│   ├── src/context/         # React Context (Auth)
│   └── src/styles/          # CSS and styling
├── config/                  # Legacy bot settings
├── modules/                 # Bot core logic (scraping, AI, resume tools)
└── docs/                    # Architecture and planning documentation

---

📋 Requirements

• Python: 3.10+
• Node.js: 18+
• Browser: Google Chrome installed (with ChromeDriver-compatible Selenium setup)
• Database: MongoDB Atlas or local MongoDB instance (Recommended for SaaS operation, falls back to local files if not set)

---

🚀 Setup & Installation

1. Configure Environment

Copy the example environment file:

Copy-Item .env.example .env

Edit .env and set the required variables:

ADMIN_USER=admin
ADMIN_PASS=change-this-password

LINKEDIN_USER=your-linkedin-email
LINKEDIN_PASS=your-linkedin-password

USE_AI=true
AI_PROVIDER=openai
LLM_API_KEY=your-api-key
LLM_MODEL=gpt-4o-mini

Optional MongoDB (Highly Recommended):

MONGODB_URI=mongodb+srv://USER:PASS@CLUSTER.mongodb.net/?retryWrites=true&w=majority
MONGODB_DB=auto_job_applier

2. Install Python Dependencies

From the project root, create a virtual environment and install backend/bot dependencies:

python -m venv .venv
.\.venv\Scripts\Activate.ps1

pip install -r requirements-bot.txt -r requirements-ui.txt

(Note: If PowerShell blocks virtualenv activation, run Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass first).

3. Install Frontend Dependencies

cd frontend
npm install
cd ..

---

🏃‍♂️ Running the Application

Development Mode

Backend: Start the FastAPI server from the project root:

.\.venv\Scripts\Activate.ps1
uvicorn server.main:app --reload --host 127.0.0.1 --port 8000

Backend URL: http://127.0.0.1:8000

Frontend: Open another terminal:

cd frontend
npm run dev

Frontend URL: http://127.0.0.1:5173 (Proxies /api requests to backend)

Production Mode

1. Build the Frontend:

cd frontend
npm run build
cd ..

2. Start the Backend Server:

.\.venv\Scripts\Activate.ps1
uvicorn server.main:app --host 0.0.0.0 --port 8000

When frontend/dist exists, server/main.py serves the built frontend directly from the FastAPI app. Production URL: http://127.0.0.1:8000

---

🤖 Onboarding Flow & Dashboard Usage

🚀 First-Time Onboarding

When you launch the web dashboard for the first time, you will be guided through a secure 5-step Onboarding Wizard to configure your automation environment:

1. Connect LinkedIn: Securely specify your LinkedIn login email and password (used by Selenium to log in and apply).
2. Upload Resume: Upload your primary resume (PDF or DOCX). You can manage multiple resumes later in the dashboard.
3. Search Rules: Define keywords (e.g. "Frontend Developer") and locations (e.g. "Remote") to target your search.
4. Configure AI: Configure OpenAI, Gemini, or Anthropic with your API key to enable smart questionnaire handling.
5. Start Automation: Complete the wizard to enter the main dashboard workspace.

📊 Dashboard & Monitoring

Once onboarding is completed, you can manage everything via the main dashboard:

• Overview: Monitor the bot's current status (Idle / Running / Errored), start or stop the bot with one click, view global statistics, and see smart setup warnings.
• Search Rules: Configure search terms, locations, experience level, and apply mode using modern chip selectors and status toggles.
• Resumes: Drag-and-drop new resumes, delete old ones, and select the default resume for active applications.
• Companies & Hiring Posts: View discovered companies and real-time scanned job posts.
• Application Logs & History: Track Selenium runs in real-time, filter application results by status, and clear history with double-confirmation popups.

Alternatively, run the bot directly from the CLI:

.\.venv\Scripts\Activate.ps1
python runAiBot.py

---

🛠 Troubleshooting

• Login uses old password: The first admin password is persisted in server/auth_state.json. Change it from the dashboard settings page, or delete the file during local development to reseed from .env.
• MongoDB not connected: If MONGODB_URI is empty or unreachable, the app automatically falls back to local file storage. Check status at http://127.0.0.1:8000/api/mongo/health.
• Chrome/Selenium errors: Ensure Chrome is up to date. If profile issues occur, set safe_mode = True inside config/settings.py.
• Resume missing: Ensure you upload and set a default resume from the dashboard Resumes tab. The app will sync the path to the bot configuration.