Where your search lives

A confidential search is only confidential if nobody else can see it. This page says, in plain language, who can. The answer is: you.

Local-first, literally

Miriam is a program that runs on your computer, not a website you log into. Your target list, your scoring criteria, every match, every note, every application you track: all of it sits in files on your machine, behind your login. There is no Miriam server holding a copy. We could not read your search if we wanted to.

When Miriam checks a company’s job board, the request goes from your machine to that company’s board, the same request your browser would make if you visited the page yourself. Nothing about it says who you are or what you’re looking for.

What leaves your machine

Nothing. Not your criteria, not your matches, not telemetry about how you use the app. There is no analytics beacon and no “anonymized usage data.” If that ever changes, it will be off by default, explained in one sentence, and your choice.

What the waitlist stores

An email address. That’s the whole record. We use it to send you an invitation when a beta seat opens, and we delete it on request: reply “remove me” to any message and it’s gone. No list is sold, shared, or enriched. Nobody who emails a waitlist for a confidential job-search tool wants mail about it; we act accordingly and send almost none.

Why it’s built this way

Most people who use Miriam are employed and would prefer their employer, their colleagues, and every recruiter on earth not know they’re looking. A hosted service would make that promise with a policy. Miriam makes it with architecture: data that never leaves your machine can’t leak, can’t be subpoenaed from us, and can’t be quietly repurposed. The policy is short because there is almost nothing to have a policy about.