Privacy Policy

When you use Atlas Job OS we collect:

• Account data — email address and hashed password (or OAuth provider token).
• CV / résumé text — extracted from documents you upload so Atlas can match jobs and draft outreach messages.
• Job search history — the searches Atlas runs on your behalf and the jobs it saves to your pipeline.
• Outreach messages — draft emails Atlas generates for your review before sending.
• Usage telemetry — token counts and cost metrics, used only for billing and service health. No message content is included.

• To operate the Atlas agent: run job searches, score matches against your CV, and draft outreach.
• To read job-related Gmail threads (recruiter replies, interview invites, application confirmations) so Atlas can track your pipeline. Atlas does not send mail on your behalf — outreach drafts are prepared in Atlas and you send them yourself from Gmail.
• To improve reliability: error logs and performance traces (via Sentry). These never include CV text or message bodies.
• We do not sell your data or use it to train AI models.

Your data is stored in a PostgreSQL database hosted on Microsoft Azure (West Europe region). CV text and agent memory files are stored as plain text files on the same server, accessible only to your user account.

• Account data is retained until you delete your account.
• Job search history and outreach drafts are retained until you delete them or close your account.
• CV text is replaced each time you upload a new CV.
• You can request full deletion at any time by emailing us.

• Google Sign-In (OAuth 2.0) — when you choose "Continue with Google" on the login or signup page, Atlas Job receives your name, email address, and profile picture from Google. We use this data only to create and identify your account. We request the minimum scopes possible (openid, email, profile) and never the ability to read, send, or modify any Google services on your behalf through this flow. Atlas does not store your Google password; authentication is handled entirely by Google. Your sign-in data is retained for as long as your Atlas Job account is active and is deleted within 30 days of account closure. You can revoke Atlas Job's access at any time via myaccount.google.com/permissions.
• Google Gmail API (metadata-only) — used only when you explicitly connect your Gmail account on the Integrations page. This is a separate, optional flow from Google Sign-In above. Atlas requests two scopes: gmail.metadata (read message headers — From, To, Subject, Date — plus the auto-generated snippet Google attaches to each message, used to detect recruiter replies, interview invites, and application status updates that match jobs in your Atlas pipeline) and gmail.labels (read your label list so Atlas can tag tracked threads with a dedicated "Atlas" label for your own inbox organisation). Atlas does not request gmail.readonly, gmail.send, gmail.compose, or gmail.modify — it cannot read full message bodies or attachments, and it cannot send, draft, delete, or alter any message in your mailbox. Gmail metadata is processed in-memory to extract job-related signals (sender, subject, snippet); only the matched metadata is persisted, never message bodies. You can disconnect Gmail at any time from Settings → Integrations, which immediately revokes Atlas's access and deletes the stored OAuth tokens. You can also revoke access independently at myaccount.google.com/permissions.
• Google Vertex AI (Gemini) — CV text and job context are sent to Gemini to power the Atlas agent. Data is processed under Google's standard API terms and is not used to train Google's models.
• Sentry — error monitoring. No personal data or message content is included in error reports.

You have the right to access, correct, export, or delete your personal data at any time. To exercise these rights, email us at Team@atlasjob.tech.

We use a single session cookie to keep you logged in. No third-party tracking or advertising cookies are used.

Questions about this policy? Team@atlasjob.tech