Job Finder

One feed that aggregates LinkedIn, Indeed, JobStreet and Google Jobs, scores every listing against your CV and explains why it matched.

Sole engineer · 2026 · Designed, built and deployed by Alexus Karl Joaquin.

What I built

• Scoring engine gives every job a 0–100 match: skill overlap (45%), title relevance (30%), recency (15%), location and remote fit (10%), with a seniority adjustment. Matched skills are shown on every card.
• Concurrent fan-out to four job boards with free direct scrapers first (SEEK JSON API, LinkedIn guest endpoint, Indeed mosaic JSON) and RapidAPI fallbacks only when a direct fetch returns nothing.
• Deduplication across sources, a 6-hour cache to respect free-tier quotas, saved / applied / dismissed tracking with optimistic UI updates.
• Upload a .docx CV and the matching profile (skills, target titles, locations) is re-extracted automatically.
• Same codebase runs as a web app and a Capacitor mobile app; backend swaps SQLite for Postgres with one env variable.

Skills demonstrated

• Python
• FastAPI
• httpx + asyncio concurrency
• Web scraping (SEEK, LinkedIn, Indeed)
• RapidAPI integration
• Scoring algorithm design
• SQLite
• PostgreSQL
• Caching & deduplication
• .docx parsing (CV extraction)
• React
• TypeScript
• Tailwind CSS 4
• TanStack Query (optimistic updates)
• Capacitor-ready mobile build
• Railway

Stack

Screens

More

All projects · Experience · LinkedIn · GitHub