8000 GitHub - thiago4int/resume-ai
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

thiago4int/resume-ai

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Make sure your ENV files are properly set with your backend, front end and ollama serving port!

🚀 AI-Powered Resume Generator

📌 Overview

This project is an AI-powered resume generator designed to take unstructured profile information and job descriptions as input, then generate ATS-friendly, structured resumes using Ollama + Gemma 3B/1B.

The goal is to transform vague, unstructured resume data into professionally formatted, job-tailored resumes, ensuring:
STAR method for experience (Situation, Task, Action, Result)
Job description alignment (matches employer expectations)
ATS compatibility (optimized for applicant tracking systems)

🛠️ Tech Stack

  • Ollama (for local LLM execution)
  • Gemma 3B/1B (lightweight LLM)
  • Node.js + Express (backend API to process resumes)
  • Handlebars.js (template engine for structured resume formatting)
  • Pandoc (for conversion to .docx)

🔧 How It Works

  1. User submits unstructured profile + job description
  2. Ollama + Gemma processes the input
    • Reformats vague experience using the STAR method
    • Aligns content with job description requirements
    • Ensures structured JSON output
  3. Handlebars.js formats the resume (Markdown → .docx conversion via Pandoc)
  4. Final resume is exported as a professional document

🚀 Installation & Setup

📌 Prerequisites

Before starting, ensure you have the following installed:
Node.js & npm
Ollama (for running Gemma locally)
Pandoc (for Markdown to .docx conversion)


🔧 Installation Steps

1️⃣ Clone the repository

git clone https://github.com/your-repo/resume-ai.git
cd resume-ai

2️⃣ Install dependencies

npm install

3️⃣ Set up environment variables Create a .env file in the root directory and add:

OLLAMA_URL=http://localhost:11434/api/generate
MODEL_NAME=gemma:3b
PORT=5001

4️⃣ Download the Ollama model (Gemma 3B/1B)

ollama pull gemma:3b

🚀 Running the API

Start the server:

npm start

It should log:

🚀 Server running on port 5001

📡 API Usage

To test the resume generator, send a request using cURL:

curl -X POST http://localhost:5001/api/resume/generate
-H "Content-Type: application/json"
-d '{ "resumeText": "John Doe is a Cloud Engineer...", "jobDescription": "Cloud Engineer - AWS, Kubernetes" }'

It should return structured JSON.

🛠️ Running Tests

1️⃣ Run Unit Tests

npm test

2️⃣ Check for Linting Errors

npm run lint

📦 Converting Markdown to .docx

After generating the Markdown resume, convert it to .docx using Pandoc:

pandoc output/resume.md -o output/resume.docx

📜 Contribution Guidelines

1️⃣ Fork the repository 2️⃣ Create a new branch for your feature 3️⃣ Submit a pull request with a description of your changes

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  
0