Privacy policy
Effective . Questions: support@peerreply.com.
The short version
- No accounts, no payments and no advertising trackers. We count visits with PostHog, without cookies, and count nothing when your browser sends Global Privacy Control or Do Not Track.
- What you type is sent to an AI model to produce your result. We do not store it; only search terms made from it are cached for up to seven days.
PeerReply is run by Maksym Shevchuk (trading as Maxan Labs), an individual entrepreneur (FOP, фізична особа-підприємець) registered in Ukraine, who decides how this data is used. Contact: support@peerreply.com.
What you type
Your notes, your paraphrase of a classmate's point, your course details, the rubric you pick and any text you ask us to check are sent to OpenAI to produce the draft or the check. We send them with storage turned off and do not save them ourselves. Under OpenAI's API terms, OpenAI may keep inputs and outputs for up to 30 days to detect abuse, unless the law requires it to keep them longer, and does not use them to train its models.
To find sources, the coach turns your notes into short search terms and sends them to PubMed (the US National Library of Medicine) and CrossRef, and looks DOIs up at doi.org. The search terms and their results are kept in our server's memory for up to seven days. Do not enter information that identifies a patient or anyone else.
If you press “Check this text on OnRubric”, the text in your “Your version” box (with the reference list, if there is one) is sent from our server to OnRubric (onrubric.com), a rubric checker run by the same team. Nothing else is sent: not the classmate's point, not your notes. OnRubric holds it for up to 10 minutes under a random code and deletes it when the link is opened; the link carries the code, never the text. If it is opened, OnRubric treats the text like any text you check there, under its own privacy policy. The code can appear in our hosting provider's and OnRubric's request logs; it works once and expires in 10 minutes. If OnRubric cannot be reached, the button opens OnRubric empty, and nothing is sent.
Our logs record the model used, its token counts and cost for each request, not what you wrote.
On your device
- rc_free, a cookie with a random id and the current day. It counts your free uses. It lasts 180 days and is renewed when the coach gives you a result.
- Your day or night choice, kept in your browser's local storage for 12 hours.
- If you turn visit counting off below, that choice, kept in your browser's local storage until you clear the site's data.
Visit counting
We count visits with PostHog (US) to see which pages and steps of the coach are used. For each page view, and when you leave a page, it receives the page address without its query (campaign labels such as utm_source excepted), the page title, the page you came from (cleaned the same way) and its site, your browser and its version, operating system, device type, language, screen and window size and time zone, whether the browser looks automated (such as a search engine's page renderer or a testing tool, judged from its automation setting, user agent and screen size), how long the page was open, and IDs that link a page view to the one before it. When you use the coach, it also receives the step and its outcome as counts and choices: the school, subject and level you picked, whether a source was searched, how many sources the draft cites, whether the coach asked questions instead and about which field, how long it took, and an error code if it failed; for a rubric check, which rubric and assignment type you picked, whether you added references and the share of rubric points awarded; which copy button you pressed, and whether you pressed the button that opens OnRubric (from a result or from the daily-limit notice) and whether it had to open OnRubric empty instead; when that button appeared under a draft and when your edit made it usable; that you started filling in the reply or rubric check form; when you follow a link to the coach from another page, that page, which tool the link opens, whether it sat in the menu or in the text, and whether its address carried more than campaign labels; and which link to NurseDraft or OnRubric you followed, from which page, and where it sat (the footer, the home page, the rubric check or the about page). It never receives what you type or what the coach writes.
No cookie is set and nothing is kept in your browser for this. Only our own browsers keep a mark in local storage that labels our visits as ours, so we can leave them out of the counts. The events go through our server, which passes on only the events, your browser's user agent and your IP address, not our cookies. PostHog tells visitors apart within a day by a one-way hash of the IP address, the user agent, the site and a salt it replaces daily, then removes the IP address and the full user agent before saving the event, so it does not store your location. PostHog keeps the events for seven years and may delete them after that.
We count nothing when your browser sends Global Privacy Control or Do Not Track, and you can turn counting off here:
On our servers
- Counters: your free uses today, keyed by the cookie's random id (kept 2 days).
- Your IP address, as part of a per-network daily counter that limits abuse (kept 2 days), and in our hosting provider's request logs.
- If you email support@peerreply.com, your message and address are kept in our Google Workspace mailbox (Gmail) so we can answer.
Who receives data
- OpenAI processes what you type to produce results.
- OnRubric (onrubric.com) receives your version only when you press the button that opens it, as described above.
- PostHog receives the visit counts described above to provide its service to us. Unless a customer opts out, its terms also let it use such data, de-identified, to develop its own products and models. Its Data Privacy Framework listing describes it as a processor of the data its customers send it.
- Vercel hosts the site and keeps request logs; Upstash runs the database for the counters.
- PubMed (NCBI), CrossRef and doi.org receive search terms and DOIs.
- Google receives and stores mail sent to support@peerreply.com in our Google Workspace mailbox (Gmail). Mail sent before November 4, 2026 came in through Resend, which may still hold copies, and was forwarded to a Gmail mailbox.
Most of these services are in the United States, so your data may be processed there. We do not sell personal data or share it for advertising.
Your rights
You can ask us what we hold about you, and ask us to correct or delete it, by writing to support@peerreply.com. If you are in the EU, the EEA or the UK, you can also complain to your data protection authority.
Children
PeerReply is for college and graduate students and is not meant for anyone under 16.
Changes
When this policy changes, the effective date at the top changes too. See also the terms of service.