Tailor — Self-hosted AI job hunting assistant

Tailor is a self-hosted, multi-agent AI assistant designed to streamline the job hunting process. It runs entirely on your machine via Docker, ensuring your sensitive data—like your resume and job preferences—never leaves your control, with the exception of necessary LLM inference calls to the Gemini API.

Screenshots

Features

• Private & Self-hosted: Your data stays local. No cloud storage, no subscriptions.
• Async Processing: Paste a job description (JD), and background agents analyze it without blocking your workflow.
• Resume-Aware: Upload your resume once and reuse it for every application.
• Smart Analysis:
  • Match Score: Get a 0-100 score based on your profile alignment.
  • Gap Analysis: Instantly see missing skills and keywords.
  • Interview Prep: Generate 5 tailored Q&As based on your specific projects and experience.
  • Communication: Auto-generate cold outreach and referral emails.
• Status Tracking: Keep track of your applications (In Progress, Applied, Rejected, Offered).
• Keyboard-Driven UI: Built for speed with vim-style navigation.

Tech Stack

Backend	Frontend	Infrastructure
Django, DRF	SvelteKit	Docker
LangGraph	shadcn-svelte	Docker Compose
Celery	Tailwind CSS	Nginx
PostgreSQL, Redis
pymupdf4llm

Multi-Agent Architecture

Tailor utilizes a multi-agent system orchestrated by a supervisor agent, with tasks handled asynchronously via Celery:

• Supervisor: Orchestrates the agent flow.
• Resume Parser: Extracts structured data from your PDF resume using pymupdf4llm.
• JD Analyzer: Extracts role, company, skills, and requirements from raw JD text.
• Gap Analyzer: Compares your resume against the JD to produce a match score and actionable recommendations.
• Question Generator: Creates 5 relevant interview questions and answers tailored to your background.
• Message Generator: Drafts professional cold emails for HR and potential internal referrals.

Self-Hosting

Tailor is built to be deployed locally using Docker.

Prerequisites

• Docker & Docker Compose
• A valid Gemini API Key

Deployment Steps

1. Clone the repository:

   git clone https://github.com/shv-ng/tailor.git
   cd tailor

2. Configure environment:

   cp .env.example .env
   # Open .env and fill in the required variables

3. Launch:

   docker compose up --build

4. Access: Open http://localhost in your browser.

Note: All your data is stored in the local PostgreSQL database. Only the prompt-formatted data is sent to the Gemini API.

Environment Variables

Variable	Description
DJANGO_SECRET_KEY	Secret key for Django
DJANGO_DEBUG	Set to False for production
POSTGRES_USER	Database username
POSTGRES_PASSWORD	Database password
POSTGRES_DB	Database name
POSTGRES_HOST	Database host
POSTGRES_PORT	Database port
REDIS_URL	Redis connection string
GEMINI_API_KEY	Your Google Gemini API key

Usage Walkthrough

1. Register/Login: Create your local account.
2. Upload Resume: Go to the Profile page and upload your resume PDF.
3. Add Job: Paste a JD into the "Add Job" form.
4. Analysis: Wait for the background agents to finish (you will see the status update).
5. Review: Open the job details to view your match score, generated questions, and emails.
6. Apply: Copy the generated emails, paste them into your preferred mail client, and hit send.