Aggregate multi-source job boards to Supabase and Google Sheets

Description

Stop manually checking dozens of career pages. This workflow runs every morning, hits the public APIs of 8+ ATS platforms and job boards, normalizes every listing into a single clean schema, and syncs everything to Supabase and Google Sheets deduplicated and ready to query.

Who it's for

Job seekers, recruiters, or career platforms that want a consolidated, up-to-date feed of openings from specific companies without scraping, without API keys, and without paying for a jobs aggregator.

How it works

A Schedule Trigger fires daily at 8 AM IST
A Company List code node defines all sources grouped by ATS type (Greenhouse, Lever, Ashby, Workable, SmartRecruiters, RemoteOK, and board APIs like Remotive, Himalayas, Arbeitnow, Jobicy)
A Prepare Request node builds the correct API URL and headers for each source, including multi-page pagination for SmartRecruiters (up to 500 jobs via offset) and Himalayas (up to 500 via page param)
An HTTP Request node fetches all sources in batches of 5
A Parse + Enrich + Filter node normalizes all divergent JSON structures into a unified schema — resolving ISO country codes, Indian city detection, salary parsing across all formats, and domain-based filtering
Deduplicated results are upserted to a Supabase (Postgres) table and written to Google Sheets

ATS platforms supported

Greenhouse, Lever, Ashby, Workable, SmartRecruiters, RemoteOK, Remotive, Himalayas, Arbeitnow, Jobicy

Normalized output schema

job_id, title, company, location, country, salary, job_type, apply_url, posted_at, source_ats

Setup

Open the Company List node and edit the sources array — add or remove companies and their ATS slugs
Update ALLOWED_DOMAINS in the Parse node to filter by location or job type relevant to you
Add your Supabase credentials in the Postgres node and confirm your table name and schema match the output fields
Connect your Google Sheets credentials and set the target spreadsheet and sheet ID
(Optional) Adjust pagination limits per source in the Prepare Request node

Requirements

Self-hosted or cloud n8n instance
Supabase project with a jobs table
Google Sheets with headers matching the normalized schema
No external API keys required — all sources use public endpoints