Settings: AI, language, data

Everything configurable lives here, including the only switch that ever lets data leave your machine: your own AI connection.

Connect your AI (optional)

  1. API mode - paste your own key for Anthropic Claude or DeepSeek. Keys are stored in the OS keychain, never in the database or logs.
  2. CLI mode - bridge a coding CLI you already subscribe to, Claude Code or Codex, instead of paying per token. This is how OpenAI models are reached: through Codex, not through an API key.
  3. A privacy note under every cloud provider states plainly what an API call sends (the job text and relevant profile fields) - see scoring for when calls happen.
Applye Settings in dark theme. AI mode is set to API (direct) with Claude (Anthropic)
            as the provider. A privacy note states that in API mode the job description and profile
            text are sent to Anthropic's servers, that the key is stored in the OS keychain and
            never written to the database or logs, and that nothing is sent until an action is
            triggered. Below it the API key section reads Stored in your OS keychain and the key
            field is empty apart from a masked placeholder.
AI settings with the privacy note.

The rest

  • Language - UI and default document language, each shown in its own name: English, Deutsch, Русский, Español, Français, Українська.
  • Appearance - dark (the default) or light, also toggleable from the top bar.
  • Job search - the geographic scope Discover filters by (worldwide / Europe / USA / Asia).
  • Delete all data - the factory reset. A two-step confirm wipes the local database and removes keys from the keychain. Deleting your data is deleting a file; there is no server copy to chase.