privacy policy

last updated: march 2026

data we collect

We collect the minimum data necessary to operate FAJAs:

  • - job listing metadata: company name, job title, location, listing age, source URL
  • - user comments and votes
  • - authentication data: provider user ID and display name from GitHub or Twitter/X
  • - vector embeddings: mathematical representations of listing content for duplicate detection

data we do NOT collect

  • - full job descriptions or listing content (never stored as text)
  • - your GitHub repositories, code, or activity
  • - your tweets, followers, or Twitter/X activity
  • - email addresses (unless provided by your OAuth provider)
  • - browsing history (though third-party ad partners may use their own tracking)

authentication

We use GitHub and Twitter/X OAuth for sign-in. This is authentication only — we request minimal permissions to verify your identity. We do not post on your behalf, read your private data, or access any platform features beyond basic profile information (user ID and display name).

cookies

We use cookies for authentication sessions and to serve ads. Third-party advertising partners may also set cookies to deliver relevant ads and measure ad performance.

vector embeddings

When a job listing is analyzed, we generate a vector embedding — a sequence of numbers that represents the mathematical “shape” of the content. This is used to detect duplicate or similar listings. The original text cannot be reconstructed from the embedding. Embeddings are stored in our database and used only for similarity search.

data storage

Data is stored in Supabase (PostgreSQL) hosted infrastructure. We do not sell, share, or provide your data to third parties except as required by law.

data deletion

To request deletion of your account or submitted data, contact us. We will remove your data within 30 days.

third-party services

  • - Supabase: database and authentication
  • - Vercel: hosting
  • - GitHub OAuth: sign-in provider
  • - Twitter/X OAuth: sign-in provider
  • - Anthropic Claude: AI-assisted metadata extraction (listing content is processed but not stored by Anthropic beyond their standard API retention policy)

changes

We may update this policy. Changes will be reflected in the “last updated” date above.

← back to fajas