Privacy · updated 18 November 2026
What happens to what you say and type
Short version: no audio or video is saved, an account is optional, your resume is never stored, and text you send to the AI is used only to give you that help.
Stays in your browser
Your saved answers with the feedback and follow-up questions they received, your stories, your score history and your theme are kept in this browser’s local storage on this device. We cannot see them. Clearing your browser data removes them; has an export button if you want a copy.
If you create an account
An account is optional. If you create one, we keep your email address, a securely hashed password, and a copy of the same data that stays in your browser (saved answers with their feedback, stories, score history and the role you are preparing for), so you can use it on another device. This is stored with Supabase, our database provider, and only you can read it. To delete your account and all of this data, email ping@mainstoryline.app from the address on your account and we will delete it. We use your email only to sign you in and to send account emails such as a password reset.
Your resume
If you add a resume on Prepare for a role, it is sent once, through our server, to Anthropic’s Claude API to draft stories and pick questions for the role you name. We do not store the resume, and it is not used to train models. Only the stories you choose to keep are saved, like any other story. Please remove details you do not want processed, such as your home address or phone number, before uploading.
Speaking practice
Your camera and microphone are used only while you press Start, and nothing is recorded. If your browser offers speech-to-text, it turns your speech into text using its own speech service (for example Google for Chrome, Apple for Safari) under that company’s privacy terms. Mainstoryline never receives your audio.
AI feedback and story help
When you ask for feedback, or ask AI to sort a story you told into four parts, the text of your answer or story (and of your follow-up answer, if you write one, and the notes of the story you selected for that question, so the feedback can point to your own details) is sent to our server and on to Anthropic’s Claude API to produce the scores. We do not store that text. Anthropic processes it as our service provider; under its commercial terms it does not use this data to train its models, and may keep it for a limited time for safety and abuse monitoring.
Please do not include other people’s names, health details or confidential business information in your answers.
Fair-use limits
To keep feedback free, we count how many feedbacks are used per day. Your browser gets a random device id (it is not linked to you), and we keep a salted, one-way hash of your IP address. These counters expire automatically within 40 days.
Hosting and logs
The site is hosted on Vercel, which keeps standard request logs (such as IP address and browser type) to run and protect the service. We do not use advertising or tracking cookies, and we do not sell or share data for advertising.
We use Vercel Web Analytics to count visits and a few actions (for example “feedback received” or “story created”). It uses no cookies, does not identify you, and never includes what you write.
Our own server also keeps anonymous daily counts (how many browsers visited, and how many practice rounds were saved, with the question id) using the random device id described above (per-browser entries expire within 40 days). These counts never include what you write, your name or your email.
Problem reports
If you report a problem with a question, we keep the question, the reason you chose and any note you write, for up to 500 recent reports. Reports do not include your answer.
Ratings and interview reports
If you rate feedback, mark a question as asked in a real interview, or tell us how an interview went, we keep what you chose or typed there: the question, your rating and reason, the scores shown on that card, the role, company, round, result, the questions you list, and how many questions and rounds you had practised. These are kept with a scrambled code for your account or browser, not your name or email, so that we can see which questions interviewers really ask and improve how answers are scored. They never include your answers or stories. We keep up to 5,000 recent records of each kind.
Marking a question as asked in a real interview needs an account. We keep one mark per account and question, the day you marked it, and any of the optional broad choices you picked (role family, years of experience, region, and roughly when you were asked), under the same scrambled code. Other visitors only see a total once at least three people have marked the same question, never who marked it. You can take a mark back at any time, and we may remove marks that look automated or duplicated.
Your choices
You can use every question without asking for AI feedback. You can delete saved answers in My Practice and stories in My Stories at any time. For any privacy question or request, contact ping@mainstoryline.app.
Mainstoryline is not intended for children under 16.