Site Icon Matthew Raynor
← Back to Services

StackJefe

Job search engine that lets developers search by the technologies they actually work with. Queries Greenhouse, Lever, Ashby, and Workable APIs to surface real listings — no login required to search. Upload a resume and Claude AI parses your skills to generate personalized outreach for any company. Power users can connect GitHub to scan orgs by tech stack, score them 0–100, and find companies before they even post jobs.

Project Overview

StackJefe is a job search engine built for developers. Search by the tech you want to work with — Django, React, TypeScript — and get real listings from 185K+ jobs across Greenhouse, Lever, Ashby, and Workable. No login required to search. Sign up with Google OAuth, upload your resume, and Claude AI extracts your skills to generate personalized outreach for any company. For power users, the GitHub org scanner finds companies building with your stack before they even post jobs — scoring them 0–100 on stack match, AI tool signals, and infrastructure maturity.

The Challenge

Job searching on LinkedIn is broken. You can't filter by tech stack, ghost jobs waste your time, and applications disappear into a black hole. Engineers deserve a search engine that speaks their language.

The Solution

Built a job search engine that queries ATS APIs (Greenhouse, Lever, Ashby, Workable) by technology. Anonymous users search freely with results blurred behind a free signup wall. Signed-up users upload resumes, get AI-parsed skills, and generate personalized outreach. GitHub org scanning finds companies before they post jobs.

Technology Stack

Backend
Django 5 Django REST Framework PostgreSQL 16 Redis Celery
Frontend
React 19 TypeScript Vite Tailwind CSS TanStack Query
Ai
Claude API (resume parsing, outreach generation)
Integrations
GitHub REST API v3 Hunter.io Apollo.io Greenhouse Lever Ashby Workable
Deployment
Railway Netlify

Key Features

Public job search — no login required, search by tech stack across 185K+ listings from 4 ATS platforms

Anonymous-to-signup funnel with blurred results (5 visible, rest behind free signup wall)

Google OAuth one-click signup with JWT cross-domain auth (Railway backend ↔ Netlify frontend)

Resume parsing via Claude AI — extracts skills, experience, and talking points automatically

Personalized outreach generation tailored to both company stack and user background

GitHub org scanner — scans dependency files, detects 200+ packages, scores orgs 0–100

ATS probing across Greenhouse, Lever, Ashby, and Workable — no API keys needed

Contact enrichment via Hunter.io and Apollo.io (BYOK model)

Business Impact

Lets developers search jobs by actual tech stack instead of keyword-stuffed LinkedIn posts

Drives signups through a low-friction anon search funnel — users convert after seeing real results

AI resume parsing + outreach generation makes direct outreach practical at scale

GitHub scanner finds companies before they post jobs — proactive, not reactive job searching

Technical Achievements

Anonymous → signup funnel that actually converts (anon search → blur wall CTA → OAuth → full results)

Cross-domain JWT auth between Railway and Netlify — clean solution to the cookie-sharing problem

Same React component serves both anon and auth users — one codebase, two experiences

Auth design separating identity (Google OAuth) from data access (GitHub OAuth) from paid services (BYOK)

Technical Implementation

Django 5 + DRF backend on Railway, React 19 + TypeScript frontend on Netlify. Google OAuth via django-allauth with JWT passed through redirect URL for cross-domain auth. The anon job search and authenticated dashboard use the same React component — visibility controlled by auth state. Celery handles background jobs (ATS scraping, GitHub scanning, resume parsing). External APIs abstracted behind provider adapters. TanStack Query for frontend polling and cache management.

Interested in This Project?

View the source code or see it in action