Local LLM LaTeX CV Tailor

An automated Python pipeline designed to customize LaTeX resumes and CVs to specific job descriptions using a lightweight local LLM (e.g., Gemma 3 1B via llama.cpp).

Unlike whole-document prompt approaches that frequently drop section titles or metadata, this tool processes experience entries company-by-company, keeping company names, job titles, and date ranges frozen in Python while rephrasing itemized bullet points to align with target job requirements.

Prerequisites

• Python: 3.8+
• LaTeX Engine: TeX Live / MiKTeX (pdflatex available in system PATH)
• Local Inference Backend: llama.cpp

Setup & Quick Start

1. Install llama.cpp & Download Model

For detailed building and installation instructions across Linux, macOS, and Windows, refer to the official llama.cpp README.

Once llama.cpp is built, create a models/ folder and download the quantized GGUF model:

cd llama.cpp
mkdir -p models
wget -O models/gemma-3-1b-it.gguf \
  [https://huggingface.co/bartowski/gemma-3-1b-it-GGUF/resolve/main/gemma-3-1b-it-Q4_K_M.gguf](https://huggingface.co/bartowski/gemma-3-1b-it-GGUF/resolve/main/gemma-3-1b-it-Q4_K_M.gguf)

2. Launch llama-server

Start the local API server backend using llama.cpp:

./llama-server \
  -m models/gemma-3-1b-it.gguf \
  --port 8013 \
  -c 4096 \
  -ngl 99 \
  --host 127.0.0.1

3. Clone Repository & Set Up Python Environment

Clone the repository and set up a Python virtual environment:

# Clone the repository
git clone [https://github.com/your-username/your-repo-name.git](https://github.com/your-username/your-repo-name.git)
cd your-repo-name

# Create virtual environment
python3 -m venv venv

Activate Virtual Environment:

Linux / macOS:

Bash

source venv/bin/activate

Windows (Command Prompt):

DOS

venv\Scripts\activate.bat

Windows (PowerShell):

PowerShell

.\venv\Scripts\Activate.ps1

4. Install Dependencies & Run Install the Python requirements and launch the pipeline:

Bash

# Install required packages
pip install -r requirements.txt

# Run the generator script
python generate_cv.py

Paste the target job description text, local text file path, or job posting URL, then press [ENTER] followed by [Ctrl+D] (or enter an empty line) to start the tailoring process.