Work03Featured
Remote Job Dev
A remote developer job board that labels every listing by where you can actually work from: worldwide, set hours, or country-restricted, with company pages, saved jobs, and application tracking.
- Role
- Fullstack
- Built for
- Personal
- Status
- active
- Stack
- Next.js, TypeScript, PostgreSQL, Python, Tailwind CSS
The project
A job board for developers looking for remote work, built around one problem: most listings just say "Remote". It aggregates 16,000+ roles from 8,000+ companies across several major job platforms, reads each posting, and labels where you can actually work from, so you only apply where you're eligible.
It has two parts: a Python aggregation pipeline that gathers, cleans, deduplicates, and classifies jobs into a Postgres database, and a Next.js web app for searching, filtering, saving, and tracking them.
Where you can work:
- Every job labeled Worldwide, Probably worldwide, Worldwide with set hours, or Location-restricted
- Labels come from the posting itself, weighing restriction phrases, listed places, time-zone requirements, and "work from anywhere" wording
- "Probably worldwide" flags jobs where only the job board claims worldwide and the posting doesn't say
- Cities and states resolved to countries, and time-zone labels normalized
Search and browse:
- Multi-select remote filter with live counts, plus location filters by continent or country that can include or exclude places
- Filters for category, experience level, job type, and source, with removable filter chips
- Company pages showing where each company hires, a breakdown of its open roles by remote type, and its most-hired roles
- Job pages with a "where you can work from" panel, key facts, and similar roles
Accounts:
- Email and password accounts with JWT sessions in httpOnly cookies
- Save jobs and track each application through saved, applied, interviewing, offer, or rejected, with a status summary
- Guests can browse the first five pages of any search before a sign-up prompt, enforced on the server
Aggregation pipeline:
- Pulls jobs from multiple sources and normalizes them into one format
- Near-duplicate detection with MinHash and locality-sensitive hashing, catching the same job reposted across cities
- Spam company blocklist, and automatic removal of listings older than 90 days
- Triggers the web app to refresh its cache after each run
Engineering:
- "Remote Ledger" design system with a custom logo, monospace data styling, and a mobile filter drawer
- Server-side caching of job queries, refreshed only when new data arrives, to keep database costs low
- Around 130 Vitest tests covering the API, auth, remote-scope labeling, guest limits, and company pages
Next project
All work