Your data, plainly.

The minimum needed to make the product work and save you time.

• Your email — to identify your account.
• Your resume — so you don't re-upload every time.
• Your voice profile — slider positions and notes, so your rewrites sound like you.
• Your tailor history — which jobs, which versions, when.
• Your credit balance and plan — so billing works.
• Your application tracking data — companies, roles, statuses, and notes you enter to track your job search.

That's the list. There is no second list.

How bullets get reframed — not who wrote them. The goal is to make the tool better for everyone.

• Anonymized transformation pairs — original bullet in, reframed bullet out. Company names, personal names, and contact info are stripped.
• Industry, seniority, confidence scores — aggregated, never individual.
• Which version gets downloaded most — refined or reframed. Helps calibrate defaults.

This is how the tool gets better — every reframe teaches the next one. If you'd rather not contribute, you can turn this off in your profile. No penalty, no nag screens.

Short version: your resume is not monetized.

• Your data is never sold.
• Your resume is never shared with third parties.
• Your resume is never used to improve the tool without stripping everything personal first. What stays is the reframe pattern, not the person.
• Your data is not kept after you delete your account.

No fine print that contradicts any of the above. That's not how this works.

• Turn off learning — in your profile settings. The tool still works the same, it just stops learning from your reframes.
• Delete your account — hard delete, everything gone, immediately. Not archived. Not “scheduled for deletion in 30 days.” Gone.
• Export your data — planned. Not built yet. Updates will be posted here.

• Encrypted at rest — Postgres on Neon, encrypted by default.
• HTTPS only — hosted on Vercel, TLS everywhere.
• Magic link authentication — no passwords stored, nothing to leak.
• Sessions expire — after 30 days, you sign in again.

Your resume is sent to external models for analysis and reframing. A variety of providers are used, routed to the best model for each task — balancing quality, speed, and cost.

Your data is subject to each provider's API terms while being processed. Only providers whose API terms prohibit training on customer data are used.

No audio or voice recordings are stored. If you use voice features, the audio is processed and discarded.