resumint-cli 🙌

resumint-cli is an AI-powered resume and cover letter generator designed to tailor your master profile to specific job descriptions in seconds. Powered by the Google Gemini API and compiled locally using Tectonic (LaTeX), it ensures your resume remains ATS-friendly, professional, and directly targeted to the role.

  ___  ___ ___ _   _ __  __ ___ _  _ _____ 
 | _ \/ __| __| | | |  \/  |_ _| \| |_   _|
 |   /\__ \ _| |_| | |\/| || || .` | | |  
 |_|_\|___/___|\___/|_|  |_|___|_|\_| |_|  

Key Features

• 🎯 AI-Powered Re-ranking & Tweaking: Automatically selects the best projects, achievements, and skills based on the Job Description (JD). Re-ranks bullet points to place the most relevant impact metrics first.
• 💻 Interactive TUI & Profile Editor: Edit your contact information and preview your master profile yaml directly within a premium dark terminal interface.
• 📝 Cover Letter Generator: Automatically drafts a professional cover letter matching your customized resume, adopting your personal tone.
• 🛠 Local LaTeX Compilation: Generates clean, type-perfect PDFs on your machine using tectonic.
• ⚙️ Self-Healing Virtual Environment: Global npm wrapper automatically sets up and maintains its own Python virtual environment.

Installation & Setup

1. Prerequisites

Make sure you have Node.js (>=16), Python (>=3.10), and the Tectonic LaTeX compiler installed on your system.

If you have Homebrew installed, you can install Tectonic using:

brew install tectonic

(If Tectonic is missing, resumint-cli will offer to auto-install it via Homebrew on first run).

2. Global Installation

Install the package globally via npm:

npm install -g resumint-cli

3. Initialize Workspace

Run the initialization command to scaffold your configuration directory in ~/.resumint:

resumint-cli init

This scaffolds the following directory structure:

~/.resumint/
├── profile.yaml          # Your master profile database (projects, skills, etc.)
├── .env                  # Contains your GEMINI_API_KEY
├── templates/            # Put custom LaTeX .tex files here to override package defaults
└── jds/                  # Directory to save job descriptions

🔑 Gemini API Key Configuration

To use the AI tailoring features, you need a Google Gemini API key:

1. Head over to Google AI Studio and create a free API key.
2. Open ~/.resumint/.env in your editor and add your key:

   GEMINI_API_KEY=your_actual_api_key_here
   GEMINI_MODEL=gemini-2.5-flash

   (Alternatively, you can export GEMINI_API_KEY in your shell profile).

🚀 How to Use

A. The Interactive TUI (Recommended)

Simply type:

resumint-cli

This launches a beautiful, tabbed terminal dashboard:

1. Tailor Resume Tab: Enter target company name, select resume type, paste the JD & any custom instruction prompts, and click Generate Resume.
2. Edit Profile Tab: Edit your personal contact details, save directly, or press Open in External Editor to open your entire profile.yaml file in your favorite terminal editor (e.g. Nano/Vim).

B. The CLI Headless Mode (Automation/Scripts)

Generate tailored resumes directly from a file:

resumint-cli --type sde --company Google --jd ~/.resumint/jds/sde_google.txt --cover

Available flags:

• --type: sde, ai, or research
• --company: Name of target company
• --jd: Path to text file containing job description/context
• --cover: (Optional) Generate matching cover letter
• --max-projects: (Optional) Limit selected projects (Default: 3)
• --max-achievements: (Optional) Limit selected achievements (Default: 5)
• --output-dir: (Optional) Custom path for output PDFs (Default: ~/resumint-output)

💡 Maximum Efficacy Guide

To get the absolute best results out of resumint-cli, follow these core tips:

1. Build a Comprehensive profile.yaml

Do not limit your profile.yaml to a single page. It is your master database!

• Add 5 to 8 projects and 10+ achievements.
• Write multiple bullet points per project (e.g. 4-6 lines) spanning different angles (e.g., frontend, performance, scaling, DB optimization).
• Gemini will select the best 3 projects and the best 3-4 bullet points per project that directly fit the target job.

2. Leverage Profile Tagging

Use tags in your profile.yaml (e.g. tags: [sde, ai, research]) to tell the generator which context projects belong to.

• For example, if you apply for an ML engineer role using --type ai, the engine prioritizes projects and achievements tagged with ai or ml, falling back to general SDE items only if needed.

3. Customize Your Personal Voice

The personal_voice block at the bottom of profile.yaml controls how the LLM drafts your cover letter and tweaks your bullet points:

personal_voice:
  tone: "Professional, confident, and metrics-oriented."
  style_guidelines: "Direct and technical. Emphasize scale, latency numbers, and design decisions."

Make this reflect your style!

4. Provide Custom Prompt Instructions

Don't just paste the Job Description. Include custom instructions in the TUI context text area or JD file:

[JOB DESCRIPTION]
... paste JD here ...

[CUSTOM INSTRUCTION]
Focus heavily on my distributed database and backend systems experience.
Make sure to emphasize Go and Kubernetes.

🎨 Customizing Templates

resumint-cli comes with professional, ATS-optimized off-campus LaTeX templates. If you want to customize the look or add university-specific formats:

1. Copy the default templates from the package directory or create your own .tex files.
2. Put them in ~/.resumint/templates/ with the names:
   • sde_offcampus.tex
   • ai_offcampus.tex
   • research_offcampus.tex
   • cover_letter.tex
3. Files placed in your home folder ~/.resumint/templates/ automatically take priority.
4. Use standard Jinja2 styling to output profile values:
   • ((( personal_info.name ))) for variables.
   • ((% for project in projects %)) / ((% endfor %)) for control blocks.

🤝 Contributing

Contributions, issues, and feature requests are welcome! Feel free to check the issues page.

License

MIT © Utkarsh Dubey