Are you a recruiter with a lot of resumes to get through and not a lot of time? Is your organization not supporting you with reasoning enabled AI advancements?

Meet AI Resume Shortlister - an intelligent tool that evaluates and ranks candidates against job descriptions using OpenAI's latest GPT-5 reasoning model. Simply put all resumes in a folder and let AI do the comprehensive analysis for you.

• GPT-5 Powered Analysis - Uses OpenAI's most advanced public reasoning model
• Beautiful Terminal UI - Rich formatting with colors, tables, and panels
• Comprehensive Evaluation - Technical skills, experience, industry relevance, education, and overall fit
• Batch Processing - Evaluate multiple resumes simultaneously
• Customizable Prompts - Modify evaluation criteria for different roles

• Add Perplexity support to concurrently research each candidate and their background.
• Evaluate accuracy / number of resumes

• OpenAI API key

1. Install UV (Python package manager)

   Visit: https://docs.astral.sh/uv/getting-started/installation/

   macOS / Linux:

   curl -LsSf https://astral.sh/uv/install.sh | sh

   Windows (PowerShell):

   powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

2. Clone and setup the project

   git clone https://github.com/pbanuru/resumeShortlister
   cd resumeShortlister
   uv sync

3. Configure environment

   cp .env.example .env.local

   Edit .env.local and add your OpenAI API key:

   OPENAI_API_KEY=sk-proj-your-openai-key-here
   

   Don't have an openai api key? Watch this video: https://www.youtube.com/watch?v=OB99E7Y1cMA

1. Add resumes - Place PDF resume files in the resumes/ folder
2. Customize job description - Edit job_description.md with your specific role requirements
3. Adjust evaluation criteria (optional) - Modify guidance_prompt.md if needed
4. Run the evaluation

   uv run main.py

The tool provides three levels of insight:

1. AI Reasoning Process - Step-by-step thinking and decision-making
2. Structured Evaluation Data - Complete JSON output with scores and reasoning for each candidate
3. Visual Summary - Beautiful tables and statistics for quick review

• Model: OpenAI GPT-5 with high reasoning effort and high verbosity
• Input Format: PDF resumes converted to base64
• Output Format: Structured JSON with Pydantic validation
• UI Library: Rich for terminal formatting
• Dependencies: OpenAI SDK, Rich, Pydantic, python-dotenv

MIT License - see LICENSE file for details.