ad
ad

Automated ATS Resume Scoring using Large Language Models and Gemini API Integration

Science & Technology

Introduction

Introduction

Hello everyone! My name is Mumar, and I am from Jaipur, Rajasthan. Currently, I am pursuing a Bachelor's degree in Computer Science Engineering from GCRC University in Jaipur, Rajasthan. I am also working as an intern at Aroga Pandit Private Limited. Today, I will be presenting my project on automated ATS (Applicant Tracking System) resume scoring using large language models (LLMs) and Gemini API integrations. This project leverages the capabilities of LLMs to improve the efficiency and accuracy of resume scoring.

Project Overview

The aim of this project is to develop an automated ATS that analyzes job descriptions along with the features of resumes. The system matches missing keywords and summarizes candidate profiles based on their resume documents. Key requirements for this project include Python 3.10, a Gemini API key, and installations to clone the project files from my account.

Development Environment

For development, I used VS Code, and the project consists of three main files. Firstly, I had to create a virtual environment using Anaconda in the project working directory. The project contains an app.py file, as well as a requirements.txt file outlining the necessary packages. Some of the packages required for this project include:

  • Google Generative AI
  • Python environment handling
  • PDF processing tools

Python Modules

In this project, I utilized various Python modules:

  • base64: For converting strings to binary format.
  • os: For interacting with the operating system.
  • pdf2image: To convert binary files into images.

The configuration of Google Generative AI involves using an API key and generating responses based on input parameters, primarily related to job descriptions and resume keywords.

Core Functionality

The project is designed to handle both PDF and Word file resumes, streamlining the application process that companies typically face when receiving numerous documents. The application provides an interface for users to upload their resumes and input job descriptions.

Key functionalities include:

  1. Uploading of resumes in both PDF and Word formats.
  2. Analyzing the resume against job descriptions to calculate a percentage match and identify missing keywords.
  3. Generating the top 10 interview questions based on the content of the resume and job description.

After uploading the job description and resume, the application evaluates the resume against the job requirements, resulting in a percentage match. For example, in a recent test case, my resume achieved a 40% match with a job description for an AI/ML role, highlighting areas where improvements could be made, such as emphasizing specific skills and experiences.

Evaluation Output

The application produces thorough evaluations, including:

  • Detailed feedback on weaknesses in alignment with the job description.
  • Recommendations for improving the resume.
  • A set of tailored interview questions based on the candidate's experience.

This tool not only assists candidates in refining their resumes but also prepares them for the interview process by providing relevant questions based on their qualifications.

Conclusion

In conclusion, the automated ATS resume scoring system designed in this project utilizes advanced natural language processing techniques to facilitate the hiring process. The integration of Gemini API enhances the model's capabilities, making it an invaluable asset for both candidates and recruiters.

Thank you for your attention!

Keywords

  • Automated ATS
  • Resume scoring
  • Large Language Models
  • Gemini API
  • Job description analysis
  • Keyword matching
  • PDF and Word files
  • Interview questions generation
  • Candidate evaluation
  • Natural Language Processing

FAQ

1. What is the purpose of the Automated ATS Resume Scoring project?
The project aims to automate the resume scoring process by analyzing job descriptions, matching keywords, and providing candidates with feedback on how to improve their resumes.

2. What technologies are used in this project?
The project utilizes Python, Google Generative AI, Gemini API, and various libraries for PDF processing and data manipulation.

3. How does the system evaluate a resume?
The system evaluates a resume by comparing its content with the job description, calculating a percentage match, and identifying missing keywords.

4. Can the application handle different file formats for resumes?
Yes, the application can handle both PDF and Word file formats for resumes.

5. How are interview questions generated?
Top interview questions are generated based on the resume content and related job description, tailored to the candidate's skills and experience.