Privacy Policy

Account information - the email address you register with, your hashed password, whether your email is confirmed.

Profile information you enter - your name, contact details (phone, mailing address, location, pronouns), the URLs of your professional profiles (LinkedIn, GitHub, personal site), your specialties, seniority, years of experience, industries, and preferred engagement track.

Documents you upload - resumes, LinkedIn exports, past employer documents, compensation documents, cover letters. Pathom parses these with the help of a large language model (see "How we use it") to pre-fill your profile and extract stories.

Opportunities you paste in - job descriptions, recruiter messages, company names, source URLs. Pathom runs evaluations on this material and generates supporting artifacts (interview prep, recruiter reply drafts, cover letters, negotiation prep, resume recommendations).

Content Pathom generates for you - evaluations, stories, resume rewrites, LinkedIn draft sections, cover letters, negotiation prep documents. These are stored on your account so you can revisit and edit them.

Compensation details - the compensation floor, current earnings, and desired range you enter, plus any benefits information.

Usage records - timestamped audit-log entries recording actions that mutate your account or that consume LLM budget (evaluation runs, resume generations, admin impersonation sessions, sign-in events). These are internal accountability records and are not shared.

Payment records (when applicable) - if you subscribe to a paid plan, we record the entitlement, ledger entries for each payment or refund, and receipts from webhook callbacks from our payment vendor. We do not directly store your credit card number; the payment vendor holds that.

To power evaluations and drafting. The core Pathom experience runs your inputs through our LLM provider (Anthropic Claude) to produce evaluations, resume rewrites, cover letters, LinkedIn drafts, and negotiation guidance. Each LLM call passes a prompt built from the content in your account combined with the specific opportunity or task you're working on.

To improve accuracy over time. We store the results of each LLM call so you can review, edit, and regenerate them with fresher inputs. We do not use your data to train the underlying model.

To operate the service. Signing you in, sending confirmation and password-reset emails, running background jobs that fire in response to your actions.

To keep the service safe and accountable. Rate limiting, abuse prevention, admin review of individual accounts when necessary, financial audit trails.

Anthropic - we use Anthropic's Claude API to generate the LLM-backed portions of Pathom's output. The content of your LLM prompts (which may include job descriptions, resume text, story bodies, and profile fields you have not opted out of sending) traverses Anthropic's systems for the duration of the API call. Per Anthropic's API terms, Anthropic does not train its models on our API inputs. See Anthropic's own privacy policy at https://www.anthropic.com/legal/privacy for their commitments.

Our payment vendor (Stripe or Paddle, depending on your region) - if you subscribe, transactional data required to process payments (email, billing country) is shared with the payment vendor. Card details are collected directly by the vendor and are not visible to us.

Nobody else. We do not sell your data. We do not share it with third-party advertisers, marketers, data brokers, or analytics providers. We do not use your data to build models we sell to anyone else.

Pathom gives you toggle-level control over which pieces of profile information Pathom is allowed to include in LLM prompts. Withholding a field means Pathom will not send it to Anthropic when generating evaluations or artifacts, at the cost of some output specificity.

Direct identifiers we never send: your full name (unless you turn it on explicitly), your email address, your phone number, and any inbound recruiter's contact information.

Fields you can toggle off: employer names, compensation figures, physical location, years of experience, patent specifics, and other content classified as semi-personally-identifiable. See your Settings page for the current defaults and the toggle for each field.

Account records - kept while your account is active. If you delete your account, we destroy the associated profile, documents, stories, opportunities, and generated content. Records tied to legal or financial obligations (payment ledger entries) are retained per applicable law.

LLM call bodies - the full prompt and response text of each LLM call is retained for approximately 90 days for debugging and regression analysis, then automatically purged. Aggregate cost accounting (token counts, timestamps) is kept longer without the body.

Payment webhook payloads - raw webhook bodies from Stripe or Paddle are retained for approximately 90 days for reconciliation, then automatically purged. The transactional records themselves (receipts, entitlement changes) are retained per applicable law.

Audit logs - internal accountability records are retained indefinitely for security and integrity purposes.

Compensation range estimates - treated as stale after 30 days and prompted for regeneration; you can regenerate at any time. Prior estimates remain viewable in your history.

See what we have - your profile page shows every field Pathom stores about you. Your dashboard shows every opportunity, story, and generated artifact.

Correct anything - every field is editable through the interface. Changes take effect immediately and flow into future LLM-generated content.

Toggle what Pathom sends to Anthropic - the Settings page carries per-field controls for the semi-personal information Pathom would otherwise include in LLM prompts.

Delete your account - the Settings page includes an account-wipe action that destroys your profile, documents, stories, opportunities, generated artifacts, GitHub sync, LinkedIn drafts, compensation data, and contact suggestions. Records retained for financial audit or legal purposes are noted in that flow.

Export your data - we plan to offer a self-serve JSON export of everything on your account. Until that ships, contact us at the address below and we will provide it.

Withdraw consent - you can revoke your agreement to this policy at any time by deleting your account. Withdrawal is prospective - content already processed cannot be retroactively unprocessed.

If you are a resident of California, Colorado, Connecticut, Utah, Virginia, or another U.S. state with a consumer privacy statute, you have specific rights beyond those listed above, including a right to obtain a copy of the personal information we hold about you and a right to appeal any refusal of a rights request. Contact us at the address below to exercise these rights. We will respond within the timeframe your state's law requires.

California residents: we do not sell your personal information as "sale" is defined by the CCPA/CPRA. We do not knowingly collect information from anyone under the age of 16. You can request a description of the personal information categories we have collected, sold, or disclosed for a business purpose within the last 12 months.

Pathom is currently designed for and operated from the United States. If you access Pathom from outside the U.S., you consent to your information being processed in the U.S., which may have data-protection standards different from your own jurisdiction. Pathom is not currently designed to meet GDPR compliance requirements and is not intended for use by EU or UK residents.

Pathom is not intended for and does not knowingly collect information from anyone under the age of 18. If we learn we have collected information from a child under 18, we will delete it.

We may update this policy from time to time. The version currently in effect is dated at the top of this page. Material changes will be surfaced in the app before they take effect; you can also check back here at any time.