JobMatcher

AI-Powered Resume to Job Matching & Analytics Platform

Overview

Developed JobMatcher, a full-stack automated pipeline that scrapes real job listings, analyzes them using AI, and provides a smart platform for users to upload their resumes. The system calculates match scores, identifies skill gaps, analyzes market demand/salary trends, and automatically generates personalized cover letters.

Project Context

This project bridges the gap between advanced RAG (Retrieval-Augmented Generation) applications and robust data engineering. It demonstrates a complete end-to-end system capable of automated continuous data collection, semantic analysis, and user-facing intelligence.

Technical Approach

  • Semantic Matchmaking (RAG): Utilized pgvector and Gemini Embeddings to enable semantic resume-to-job matching, surpassing basic keyword search
  • Automated Cloud Scraping Pipeline: Engineered Playwright-based scraping with advanced Anti-Detection techniques (stealth mode, UA rotation), automated via GitHub Actions
  • Robust ETL Workflow: Extracted and structured job data (skills, experience) using the Groq API, and generated semantic embeddings with Google Gemini
  • Market Analytics Dashboard: Visualized real-time tech skill demands and salary trends mapped against specific tools and technologies

Key Features

  • Skill Gap Analysis: Precisely highlights which required skills from a job description are missing from an applicant’s resume
  • Cover Letter Generation: Automatically drafts highly tailored cover letters referencing specific job requirements and user experience
  • Data Freshness: Fully automated nightly scraping ensures the platform’s job market analytics represent current industry realities

Technologies Used

Category Tools
Frontend Next.js 15 (React), TailwindCSS, Recharts
Backend & DB Python, FastAPI, Supabase (PostgreSQL + pgvector)
AI / LLMs Google Gemini (Embeddings/Generation), Groq (Fast JSON Extraction)
Automation GitHub Actions, Playwright

Source & Links