Privacy

LimeRoost is a product of Sur Labs, LLC, a Delaware company in the United States, registered at 131 Continental Dr, Suite 305, Newark, DE 19713. The product is operated from Argentina. For anything about privacy, including deleting your data, write to hola@dametrabajo.com.

What we store is what you enter or what you generate using the product:

• Your account: email address and password. The password is stored encrypted and we cannot read it, neither we nor anyone else.
• Your professional profile: name, location, years of experience, education, languages, work history and whatever you loaded from your CV.
• Your search preferences: which countries can hire you, how you want to work, the salary figures you set and the list of things you asked us never to claim.
• Your applications: the postings you saved, the status of each one and the notes you wrote.
• The generated documents: the CVs and cover letters built for each posting.
• Your personal brand content: the scheduled texts and, if you pasted them, the links and the numbers of what you published.
• The replies that arrive at your LimeRoost address, if you set one up: who wrote, the subject and the text of the message.

That last point deserves a paragraph of its own, because it is your correspondence. The LimeRoost address forwards everything to your usual inbox — that has not changed — and it now also stores the message, so you can read it again inside the application, next to the application it belongs to. The text of the message is deleted twelve months after it arrives, automatically. What remains is the row — who it was from, with what subject, about which posting — which is what makes the history of a search readable, and that is not the correspondence.

Salary figures and interview notes are among the most sensitive things in a job search. They are treated as such: they live in your account and they do not leave it.

All of that sits in a database on our own server, rented from an infrastructure provider, not on a third-party platform that also does something else with the data. The database has row-level access rules: your account can only read and write its own rows. It is not a promise that we will not look, it is that the database rejects the query.

There is one exception worth stating plainly: as administrators we have technical access to the server and the database, which is what it takes to make backups and fix things when they break. We do not use that access to read anyone's data, unless you ask us for help with something specific.

We do not sell data and there is no candidate showcase. Your profile does not appear in search engines and nobody finds you here. Nothing on this list — your profile, your CV, your applications, your salaries, your messages — is shared with anyone for advertising, with or without permission. The only thing that can go out for that is what we explain under Measurement, and only if you accepted it.

The extension exists so you can save postings without copying them by hand. It works like this, and it does nothing else:

• It runs on job posting pages on LinkedIn and on the portals in the list, on your LinkedIn feed, on the page of a post you opened, on the results page of your own posts, and on the LimeRoost application. It does not activate on any other page.
• When you press the save button, it reads what is on screen: title, company, location, work mode, the text, the link and whether the site marked it as already applied.
• That stays in your own browser's storage, on your computer. It is not sent to any server.
• It only travels to your account when you open the application and press import. From there it lives in your account, with the same access rules as everything else.

There are two more things it can read, and you have to ask for both:

• A LinkedIn post that is a job offer, whether you opened it or saw it go by in your feed. Plenty of offers circulate as posts and not as postings. It reads the text of that post, who wrote it and the links it carries, and sends it to your account only when you press save.
• The numbers of your own posts — impressions, reactions, comments — on the page where LinkedIn shows them to you. Nobody else's are read, and no company name or audience region is stored: only how much the first of each list weighs.

The extension does not read your messages, your contacts or your profile. It does not inject code fetched from the internet, it does not use tracking cookies, it shows no advertising and it sends no data to third parties. Nothing it reads leaves your browser until you press save.

When you install it, Chrome warns that it can read and change your data on linkedin.com, without saying on which pages. That is the browser's wording: permissions are granted per domain, not per screen, and there is no way to request only the six pages listed above. It is needed because LinkedIn changes screens without reloading, and without that permission the save button never comes back when you move from one posting to the next. What is read on each page is written above, and the extension's code can be reviewed in full: it is the same code that gets installed.

To delete everything it saved, just empty the queue from its own menu, or uninstall it.

This changed, and it gets its own section because the previous version of this page promised the opposite: that the feed was never read. Now it is looked at, and it is worth being exact about what that means.

When you are on your feed, the extension looks at the text of the posts already drawn on your screen to decide whether any of them is a job offer and give it a save button. Many of the best offers in the region circulate that way and not as postings. That text does not leave your browser: the post is sent to your account only when you press the button.

And what it does not do, which is the part that matters: it does not walk your feed on its own, it does not scroll automatically, it does not ask LinkedIn for anything your browser has not already asked for, and it stores absolutely nothing from posts you did not choose to save. That is the difference between a tool and a scraper, and it is the reason this can exist without putting your account at risk.

When you generate a CV or a cover letter for a posting, the text of the posting and the parts of your profile that are needed are processed by a language model from an external provider, which returns the draft. The providers we use do not train their models on what is sent to them through their programming interface.

We never send them your password, your private interview notes, or the links to your other applications. And if you turned on interview practice, the audio of that call is processed on the spot so it can answer you, and it is not stored.

If you use the extension's panel assistant, it receives the text of the posting on your screen and a summary of your profile and how your search is going, which is what lets it answer about that specific role and not in general. It does not receive your password or your private notes.

Of that conversation, only a counter stays on our server: how many messages you used today and when the last one was, which is what holds up the daily cap. That counter is deleted after seven days. The conversation itself is stored in your own browser's storage, not in your account, so uninstalling the extension takes it away.

A different matter is the dictation for the personal brand questions, which is optional and has its own button. It uses the speech recognition your browser already ships with: Chrome does not resolve it on your machine, it sends the audio to Google and returns the text. It does not pass through our servers, we only receive whatever ended up written in the box. If you would rather your voice did not leave your computer, type the answer: for us it is exactly the same.

We write to you about the product: the welcome email, a weekly summary of your search and important notices about your account. We send them with Resend, which processes your email address for that.

The weekly summary can be stopped whenever you want, with the link at the bottom of that same email. Account notices, like a password change, cannot be stopped because they are part of the service.

If you turn it on, we give you your own address ending in @foliomail.me to use in your applications. It exists so replies all land in one place and so the board moves on its own when a company answers you.

What happens with what arrives there, said in full: the body of the message is stored in your account, a language model reads it to decide whether it is an interview invitation, a rejection or an automated notice, and it is forwarded to your usual inbox. Attachments travel with that forward exactly as they arrived.

That message was written by another person, so there is data of theirs inside too: their name, their address and whatever they wrote. We keep it because the process it belongs to is yours, we do not use it for anything else, and we show it to nobody outside your account.

It is deleted with your account, like everything else. And if you would rather not have the address, do not turn it on: the whole product works without it.

We count visits to this site with our own tool, hosted on our own server. That one uses no cookies and builds no profile of you: it counts page views and which site you came from, and none of that identifies you or follows you anywhere else. It always runs and we ask you nothing, because there is nothing to ask.

Separately from that we run advertising, and there third parties are involved. When you arrive from an ad on Meta, LinkedIn or Reddit, those platforms want to know whether the ad worked, and for that they set a cookie. We ask you about that with a banner the first time, and we load nothing until you say yes. If you say no, the site works exactly the same and we keep counting visits with our own tool.

Two clarifications that matter more than they look. First: this happens on limeroost.com and not inside the application. On app.limeroost.com there is no advertising pixel at all, so neither Meta nor LinkedIn nor Reddit sees which postings you look at, what salary you set or which country you want to move to. Second: if you sign up or pay, and only if you accepted the banner, we tell that platform from our server that the person who clicked its ad ended up converting. Your email goes irreversibly hashed, which is how those platforms match the conversion, and nothing from your profile, your CV or your applications goes with it.

If you said no, we do not send that either. The answer you gave in the banner travels with you all the way to the server and is what decides, not a separate checkbox.

The report-a-bug button sends your comment, the section you were on, the last technical errors on the page and, if you took one, a screenshot. You take the screenshot: the browser asks you first what you want to share, and you can send none.

We do read that, which is the whole point of a report. Keep in mind that if the screenshot shows data from your screen, that data goes in the report.

Most errors are reported by nobody: they happen, the person tries again and carries on. To be able to fix them we use two tools inside the application, and it is worth saying exactly what they see.

• Sentry receives the technical error: the message, which line of the program it happened on and which screen you were using. We strip the email, the text of the fields and anything you typed before sending it.
• LogRocket records how the screen moved, so we can see what led to the error. What is recorded is the shape of the page, not its content: all text and everything you type comes out masked, so in the recording your CV, your salaries, your messages and your notes look like grey blocks.

Both run only inside the application and not on this site. Both are third-party services, in the United States, and both are instructed to delete what they receive after thirty days.

If you would rather we did not record you, write to hola@dametrabajo.com and we turn it off for your account. It affects nothing else in the product.

There are three ways to pay and none of them puts your card details through our servers. Of your subscription we store which plan you have, whether it is up to date and until when, and the last four digits of the card when the payment method returns them — so you recognise the charge on your statement, and for no other reason.

Mercado Pago, in Argentine pesos: card details are entered on their side.

Paddle, in dollars: here Paddle is not just the processor, it is the seller of record for the subscription. Your billing data belongs to them and is governed by their privacy policy; what reaches us is that you paid, which plan and until when.

USDC, in cryptocurrency: the transfer goes to an address of ours on the Solana network. Bear in mind that that is public: anyone looking at the network sees the amount and both addresses. We do not know whose wallet it is unless you tell us when claiming the payment.

As long as you have an account. If you delete it, everything of yours goes with it: profile, applications, documents and content. No copy of ours remains, beyond the server backups, which rotate on their own and are overwritten within a month.

You close the account yourself, from Profile, and it is deleted on the spot. You can also write to hola@dametrabajo.com from your account address to ask us for a copy of everything we hold about you, to correct something that is wrong, or to stop writing to you.

Everything of yours is deleted. What remains is one ownerless line: the month you started, how many times you applied, how many replies you got, whether you found a job and after how many weeks, which plan you had and the country you declared.

It carries no email, no name, and no identifier that could lead back to you. We keep the month and not the day precisely for that reason: an exact date, crossed with anything else, points at a person again.

We do it for a concrete reason worth stating: the people who find a job are the ones who cancel, so without this the only people whose story we could tell would be those who did not. It is what lets us know — and say with numbers — whether the product works. We ask whether you found a job when you close the account, and answering is optional.

The product is for people over 18 and is not intended for minors. We do not ask for or seek data about minors.

When something substantial changes, the date above changes and we tell you by email before it applies. We do not change the rules retroactively and in silence.