CV Generator   

A tool to generate formatted, tailored PDF CVs from a simple text file.

 

📦 TOML input  →  ⚙️ Python + LaTeX  →  📄 PDF output

 

✨ About The Project

I got tired of juggling multiple CV versions — academic, industry, grants — and constantly editing .tex files. So, I built this tool to separate content (your info) from design (the LaTeX template) and make CV updates easy.

• Using a single TOML (toml?) file plus facets, you can quickly create different CV versions with one command.
• Docker support means no LaTeX install is needed.

---

Getting Started

Prerequisites

• Python (3.10+)
• uv (optional but recommended)
• Docker Desktop (optional but recommended)

Installation

Clone the repository:

git clone https://github.com/your-username/cv-generator.git
cd cv-generator

Install dependencies:

# with uv (recommended) 
uv sync

# or:
# pip install .

---

Usage

Build Your CV

# (recommended)
uv run task build

# or:
# python app/main.py

Clean the Output Directory

To remove generated files:

# (recommended)
uv run task purge

# or: 
# python app/scripts/clear_output.py

---

⚙️ How It Works

This system revolves around three main components:

1. Content File

Located in app/content/ is a TOML file that stores your personal details, experience, education, and skills.

Each entry can include facets to generate different versions of the CV — for example, targeting academia vs. industry:

[[education]]
degree = "Doctor of Science in Physics"
institution = "University of Paris (Sorbonne)"
# first facet
details.academia = [ 
    "Thesis established the new scientific field of radioactivity..."
]
# second facet
details.industry = [
    "Discovered and characterized new materials (Polonium, Radium)..."
]

👉 See the full example: mariecurie_cv.toml

The active facet is selected in the config file.

2. Configuration

The config.toml file is the control center for the build process.

Key sections:

Section	Key	Description
[cv_settings]	facet_to_generate	Select which version of the CV to generate
	content_file_name	Path to the TOML content file
	template_file_name	Path to the Jinja2-LaTeX template
[build_settings]	use_docker	Build with Docker or local LaTeX
[output_settings]	output_format	pdf (default) or tex (for LaTeX source only)
	cleanup_temp_files	Remove intermediate files like .aux, .log, etc.
	output_mode	auto (timestamped folder) or dump (flat output)
	base_filename	Custom output filename (e.g. CV_MarieCurie_2025)

3. Templates

Template files are located in /app/templates/ and define the overall layout of the CV using LaTeX + Jinja2. These templates are populated with the content defined in your TOML file.

• cv_template.tex.j2: The main document structure.

• partials/: Modular components for specific sections such as:

  • _header.tex.j2
  • _skills.tex.j2
  • _experience.tex.j2
  • _education.tex.j2
  • ...

Together, these templates control the visual style, structure, and content placement in the final document.

---

🐳 Docker Workflow

Use Docker for consistent builds across systems and to avoid any local LaTeX setup or dependency issues.

To use docker, simply:

1. In config.toml, set: use_docker = true
2. Start the build (uv run task build)

Under the hood this will:

• Automatically build the Docker image
• Mount your project into the container
• Compile your CV to PDF or LaTeX

📂 Project Structure

.
├── app/
│   ├── content/         # Your CV data files (.toml)
│   ├── templates/       # Jinja2 LaTeX templates (.j2)
│   ├── assets/          # Static assets like a signature image
│   ├── scripts/         # Helper Python scripts
│   ├── config.toml      # Main configuration file
│   └── main.py          # The orchestrator script
├── output/              # Generated CVs appear here
├── Dockerfile           # Defines the Docker build environment
├── pyproject.toml       # Project definition and dependencies
└── README.md            # You are here!

---

📄 License

Distributed under the MIT License.