README.md

Name Parser

Parses names into first name, last name, and title using AI.

Overview

The Name Parser uses the Groq API (Llama 3) to intelligently analyze names and split them into their components - even for complex cases like academic titles or noble predicates.

Requirements

• Python 3.10+
• Groq API Key (free at https://console.groq.com/keys)

Installation

pip install pipenv
pipenv install
pipenv shell

Configure API Key

cp .env.example .env
# Edit .env and add your GROQ_API_KEY

Usage

python src/main.py "Dr. Max Mustermann"

Output:

Parsing: Dr. Max Mustermann
Model:   llama-3.1-8b-instant

First name: Max
Last name:  Mustermann
Title:      Dr.

More Examples

python src/main.py "Prof. Dr. Anna Maria Schmidt"
python src/main.py "Hans Müller"
python src/main.py "von Goethe, Johann Wolfgang"

Custom Configuration

python src/main.py "Dr. Max Mustermann" -c my_config.yaml

Configuration

Edit config/settings.yaml to customize the LLM model and prompt:

# LLM Settings
llm:
  model: "llama-3.1-8b-instant"
  temperature: 0.1
  max_tokens: 200

# System prompt for the LLM
prompt: |
  You are an expert in name analysis.
  Parse names into their components: first name, last name, and title.
  ...

Project Structure

name_parser/
├── src/
│   ├── __init__.py
│   ├── name_parser.py   # AI-based name parsing
│   └── main.py          # CLI entry point
├── config/
│   └── settings.yaml    # Configuration (model, prompt)
├── logs/
│   └── requests.log     # Request logging (gitignored)
├── Pipfile              # Dependencies
├── .env.example         # API key template
└── README.md

Logging

All requests are logged to logs/requests.log with:

• Timestamp
• Git commit hash
• Model used
• Input and parsed output

License

MIT License