Resume Matcher Setup Guide

English | Español | 简体中文 | 日本語

Welcome! This guide will walk you through setting up Resume Matcher on your local machine. Whether you're a developer looking to contribute or someone who wants to run the application locally, this guide has you covered.

Table of Contents

• Prerequisites
• Quick Start
• Step-by-Step Setup
  • 1. Clone the Repository
  • 2. Backend Setup
  • 3. Frontend Setup
• Configuring Your AI Provider
  • Option A: Cloud Providers
  • Option B: Local AI with Ollama
• Docker Deployment
• Accessing the Application
• Common Commands Reference
• Troubleshooting
• Project Structure Overview
• Getting Help

Prerequisites

Before you begin, make sure you have the following installed on your system:

Installing uv (Python Package Manager)

Resume Matcher uses uv for fast, reliable Python dependency management. Install it with:

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# Or via pip
pip install uv

Quick Start

If you're familiar with development tools and want to get running quickly:

# 1. Clone the repository
git clone https://github.com/srbhr/Resume-Matcher.git
cd Resume-Matcher

# 2. Start the backend (Terminal 1)
cd apps/backend
cp .env.example .env        # Create config from template
uv sync                      # Install Python dependencies
uv run uvicorn app.main:app --reload --port 8000

# 3. Start the frontend (Terminal 2)
cd apps/frontend
npm install                  # Install Node.js dependencies
npm run dev                  # Start the dev server

Open your browser to http://localhost:3000 and you're ready to go!

Note: You'll need to configure an AI provider before using the app. See Configuring Your AI Provider below.

Step-by-Step Setup

1. Clone the Repository

First, get the code on your machine:

git clone https://github.com/srbhr/Resume-Matcher.git
cd Resume-Matcher

2. Backend Setup

The backend is a Python FastAPI application that handles AI processing, resume parsing, and data storage.

Navigate to the backend directory

cd apps/backend

Create your environment file

cp .env.example .env

Edit the .env file with your preferred text editor

# macOS/Linux
nano .env

# Or use any editor you prefer
code .env   # VS Code

The most important setting is your AI provider. Here's a minimal configuration for OpenAI:

LLM_PROVIDER=openai
LLM_MODEL=gpt-5-nano-2025-08-07
LLM_API_KEY=sk-your-api-key-here

# Keep these as default for local development
HOST=0.0.0.0
PORT=8000
FRONTEND_BASE_URL=http://localhost:3000
CORS_ORIGINS=["http://localhost:3000", "http://127.0.0.1:3000"]

Install Python dependencies

uv sync

This creates a virtual environment and installs all required packages.

Start the backend server

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

You should see output like:

INFO:     Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
INFO:     Started reloader process

Keep this terminal running and open a new terminal for the frontend.

3. Frontend Setup

The frontend is a Next.js application that provides the user interface.

Navigate to the frontend directory

cd apps/frontend

(Optional) Create a frontend environment file

This is only needed if your backend runs on a different port:

cp .env.sample .env.local

Install Node.js dependencies

npm install

Start the development server

npm run dev

You should see:

▲ Next.js 16.x.x (Turbopack)
- Local:        http://localhost:3000

Open http://localhost:3000 in your browser. You should see the Resume Matcher dashboard!

Configuring Your AI Provider

Resume Matcher supports multiple AI providers. You can configure your provider through the Settings page in the app, or by editing the backend .env file.

Option A: Cloud Providers

Example .env for Anthropic:

LLM_PROVIDER=anthropic
LLM_MODEL=claude-haiku-4-5-20251001
LLM_API_KEY=sk-ant-your-key-here

Option B: Local AI with Ollama (Free)

Want to run AI models locally without API costs? Use Ollama!

Step 1: Install Ollama

Download and install from ollama.com

Step 2: Pull a model

ollama pull gemma3:4b

Other good options: llama3.2, mistral, codellama, neural-chat

Step 3: Configure your .env

LLM_PROVIDER=ollama
LLM_MODEL=gemma3:4b
LLM_API_BASE=http://localhost:11434
# LLM_API_KEY is not needed for Ollama

Step 4: Make sure Ollama is running

ollama serve

Ollama typically starts automatically after installation.

Docker Deployment

Prefer containerized deployment? Resume Matcher includes Docker support.

Quick Start with Docker Compose

# Start the container from a published image
docker compose up -d

# View logs
docker compose logs -f

# Stop the container
docker compose down

Customizing Ports

# Change host port only (container stays on 3000)
PORT=4000 docker compose up -d

Configuration Options

Note: Changes to LOG_LEVEL and LOG_LLM require a container restart to take effect.

Using Ollama with Docker

To use Ollama running on your host machine:

LLM_API_BASE=http://host.docker.internal:11434 docker compose up -d

Then configure Ollama as your provider in the Settings UI.

Using Docker Secrets

The container supports *_FILE from docker secrets. For sensitive values, you can mount a secret file and point to it:

LLM_API_KEY_FILE=/run/secrets/llm_api_key docker compose up -d

Supported *_FILE variables:

Rules:

• Use either the variable or its *_FILE variant, not both.
• If both are set, the container exits with an explicit error.

Logging Level Configuration

You can tune logs globally and for LiteLLM separately:

LOG_LEVEL=INFO LOG_LLM=DEBUG docker compose up -d

Security warning: LOG_LLM=DEBUG causes LiteLLM to log API keys in plaintext. Do not use DEBUG level in production or shared environments. The default WARNING is safe.

Note: LiteLLM also reads the LITELLM_LOG environment variable internally to control handler-level filtering. LOG_LLM sets the logger level. Both must allow a message for it to appear. If you set LITELLM_LOG from LiteLLM docs, make sure LOG_LLM is set to an equal or lower level.

Important Notes

• API keys are best configured through the UI at http://localhost:3000/settings
• Data is persisted in a Docker volume (resume-data)
• The Settings UI configuration is stored in the volume and persists across restarts
• App and API share the same origin: frontend on /, API on /api

Accessing the Application

Once the container is running, open your browser:

First-Time Setup Checklist

1. Open http://localhost:3000/settings
2. Select your AI provider
3. Enter your API key (or configure Ollama)
4. Click "Save Configuration"
5. Click "Test Connection" to verify it works
6. Return to Dashboard and upload your first resume!

Common Commands Reference

Backend Commands

cd apps/backend

# Start development server (with auto-reload)
uv run uvicorn app.main:app --reload --port 8000

# Start production server
uv run uvicorn app.main:app --host 0.0.0.0 --port 8000

# Install dependencies
uv sync

# Install with dev dependencies (for testing)
uv sync --group dev

# Run tests
uv run pytest

# Check if database needs reset (stored as JSON files)
ls -la data/

Frontend Commands

cd apps/frontend

# Start development server (with Turbopack for fast refresh)
npm run dev

# Build for production
npm run build

# Start production server
npm run start

# Run linter
npm run lint

# Format code with Prettier
npm run format

# Run on a different port
npm run dev -- -p 3001

Database Management

Resume Matcher uses TinyDB (JSON file storage). All data is in apps/backend/data/:

# View database files
ls apps/backend/data/

# Backup your data
cp -r apps/backend/data apps/backend/data-backup

# Reset everything (start fresh)
rm -rf apps/backend/data

Troubleshooting

Backend won't start

Error: ModuleNotFoundError

Make sure you're running with uv:

uv run uvicorn app.main:app --reload

Error: LLM_API_KEY not configured

Check your .env file has a valid API key for your chosen provider.

Frontend won't start

Error: ECONNREFUSED when loading pages

The backend isn't running. Start it first:

cd apps/backend && uv run uvicorn app.main:app --reload

Error: Build or TypeScript errors

Clear the Next.js cache:

rm -rf apps/frontend/.next
npm run dev

PDF Download fails

Error: Cannot connect to frontend for PDF generation

Your backend can't reach the frontend. Check:

1. Frontend is running
2. FRONTEND_BASE_URL in .env matches your frontend URL
3. CORS_ORIGINS includes your frontend URL

If frontend runs on port 3001:

FRONTEND_BASE_URL=http://localhost:3001
CORS_ORIGINS=["http://localhost:3001", "http://127.0.0.1:3001"]

Ollama connection fails

Error: Connection refused to localhost:11434

1. Check Ollama is running: ollama list
2. Start Ollama if needed: ollama serve
3. Make sure the model is downloaded: ollama pull gemma3:4b

Project Structure Overview

Resume-Matcher/
├── apps/
│   ├── backend/                 # Python FastAPI backend
│   │   ├── app/
│   │   │   ├── main.py          # Application entry point
│   │   │   ├── config.py        # Environment configuration
│   │   │   ├── database.py      # TinyDB wrapper
│   │   │   ├── llm.py           # AI provider integration
│   │   │   ├── routers/         # API endpoints
│   │   │   ├── services/        # Business logic
│   │   │   ├── schemas/         # Data models
│   │   │   └── prompts/         # LLM prompt templates
│   │   ├── data/                # Database storage (auto-created)
│   │   ├── .env.example         # Environment template
│   │   └── pyproject.toml       # Python dependencies
│   │
│   └── frontend/                # Next.js React frontend
│       ├── app/                 # Pages (dashboard, builder, etc.)
│       ├── components/          # Reusable React components
│       ├── lib/                 # Utilities and API client
│       ├── .env.sample          # Environment template
│       └── package.json         # Node.js dependencies
│
├── docs/                        # Additional documentation
├── docker-compose.yml           # Docker configuration
├── Dockerfile                   # Container build instructions
└── README.md                    # Project overview

Getting Help

Stuck? Here are your options:

• Discord Community: dsc.gg/resume-matcher - Active community for questions and discussions
• GitHub Issues: Open an issue for bugs or feature requests
• Documentation: Check the docs/agent/ folder for detailed guides

Useful Documentation

Happy resume building! If you find Resume Matcher helpful, consider starring the repo and joining our Discord.