A zero-training policy. A 24-hour wipe button with a deletion receipt. A compliance roadmap with real dates, not theatre.
We are pre-SOC 2. Rather than name an auditor we have not yet engaged, every report below shows its real status today and the quarter we plan to start the work. We will update this page the day we sign an audit letter.
It is our policy that your resume bullets, interview transcripts, and emails are not used to train any model — ours or our sub-processors'. We route LLM inference through providers' zero-retention tiers where available. Enterprise customers can request this commitment in writing in a DPA. If we ever change this policy, we will give notice before it takes effect.
When enabled, we strip name, email, phone, address, school, employer and any free-text PII from the payload before it ever touches an LLM. The model sees role / skills / outcomes. You see the final draft un-redacted. This survives across resume, interview copilot, outreach drafting.
Settings → Privacy Vault → Wipe everything. We start a 24-hour countdown (so accidental clicks are recoverable). At T+24h every row tied to your account is hard-deleted from the primary DB; replicas and encrypted backups roll off within 30 days. You receive a deletion receipt listing what was purged and when. A cryptographically-signed receipt format is on our roadmap — see below for the shape we are building toward.
Today our application + database run in a single US region. We intend to honour GDPR + UK-GDPR data-subject rights listed below regardless of where you live. A dedicated EU region for EEA / Swiss / UK customers is on the roadmap — we will not claim residency we have not shipped.
Sensitive vault contents — resume source profile, interview recordings, and offer letters — are encrypted at rest. We are building toward a per-user key derived from your password so that even our operators cannot read them; until that ships, access is restricted by strict internal controls and audit logging. We will keep this section accurate as we roll it out.
Our hosting provider (compute + storage), our database provider, Anthropic (LLM inference — we use zero-retention tiers where available), Stripe (billing), and our email provider. Anyone who touches your data is on this list, and we keep it current. We aim to give 30 days notice before adding a sub-processor.
Most “delete my account” buttons just flip a flag. Ours initiates a real, audit-logged purge — with a 24-hour grace window so you don’t lose your search to a fat-fingered tap.
{
"version": "1.0",
"issued_at": "2026-05-29T14:02:11Z",
"subject": {
"user_id": "usr_8XK4...",
"email_hash": "sha256:f93c..."
},
"purged": {
"users": 1,
"source_profiles": 1,
"resumes": 7,
"applications": 41,
"contacts": 18,
"interview_sessions": 6,
"offers": 2
},
"merkle_root": "0x9d7a4f1c...",
"issuer_pubkey": "ed25519:maxapply-2026",
"signature": "ed25519:0x4a1b..."
}Settings → Export. We email you a machine-readable archive of every record we hold on you, typically within 24 hours.
Settings → Privacy Vault → Wipe. 24-hour countdown, then hard-delete + receipt. No 'are you sure' loops.
Email privacy@maxapply.ai. We acknowledge within 72 hours and action within 30 days, per GDPR.
We collect only what we need to run the product: your account email, the resume / interview / offer data you give us, and basic usage telemetry to keep the service up. We never sell it, never use it to train a model (see the zero-training policy above), and never share it with anyone outside the sub-processors listed on this page.
Questions or a data-subject request? Email privacy@maxapply.ai.
Use MaxApply AI to find work, sharpen your materials, and negotiate offers — not to break the law or other people’s terms. We provide the service “as is” while it is in active development and will give notice before material changes.
Full terms or an enterprise agreement? Email legal@maxapply.ai.
We respond to every email at security@maxapply.ai within one business day. PGP key on request.