What we collect, why, and what we do not do.

lexica is built by one person (Railly Hugo). The data we collect is the minimum needed to run a vocabulary app: who you are (Clerk handles auth), what words you have rated, what you have practiced, and what device you used.

We do not sell data, do not track you across the internet, do not run third-party advertising scripts.

Account: email, name, and any social login info you connect via Clerk. Stored in Clerk, not on our servers.

Preferences: your native language, practice language, translation language. Stored in our Postgres database.

Practice data: your assessment results, lemma ratings (solid / shaky / unknown), nugget progress, and which cards you have reviewed. Stored in our Postgres database keyed to your Clerk user ID.

Waitlist: email and (optional) the role or domain you indicated when joining. Stored by Clerk's Waitlist feature.

Analytics: aggregate, anonymous page views and named events (such as joining the waitlist or opening a domain) via Vercel Web Analytics. No cookies, no third-party tracking pixels, no IP storage.

To make spaced repetition work, we need to remember which cards you have seen and how you rated them. That is the entire product.

To prioritize which vertical tracks to ship next, we count waitlist votes per domain.

To know which features actually matter, we look at aggregate event counts. We do not look at individual users.

Clerk (auth, waitlist) · clerk.com

Neon (Postgres database) · neon.tech

Vercel (hosting, analytics) · vercel.com

Cloudflare R2 (audio and image assets) · cloudflare.com

OpenAI, Anthropic, ElevenLabs (used by us to generate content; we do not send your personal data to them).

Export your data: every nugget you have reviewed can be exported to Anki TSV any time. Settings page lets you download a JSON dump of everything else.

Delete your account: contact us and we wipe your row from our database within seven days. Clerk handles the auth side of deletion.

Correct your data: change your preferences in the app any time. Email us for anything else.

lexica is for adult learners (B2 English and up). We do not knowingly collect data from anyone under 13. If you believe we have, contact us and we will delete it.

If we change anything material in this policy we will email everyone with an account before the change takes effect. The git history of this page is public on GitHub.

Questions: privacy@railly.dev (or reply to any lexica email). We answer within a few days.