PrivacyUpdated 22 October 2026

The short version

A Praximedes runs on your machine, and what it learns about your work stays with you. I do not keep a copy of it, and nothing one person grows is used to train anything for anyone else.

This site keeps only what I need to answer you, to sign you in, and to bill you. Nothing is sold, and nothing here follows you around the web.

The rest of this page is the long version: every field, every company that touches it, how long it stays, and what you can ask me to do with it.

Who holds it

I am Ludmilla, and I run Praximedes from Lisbon, Portugal. The company that is legally responsible for the data is named under Legal details, at the end of this page.

There is no separate data protection officer. Questions, and every request about your data, come to me at hello@praximedes.com.

What the site keeps

Exactly these fields, and nothing else, at each moment.

When you write to me: Your name, your email address, what you wrote, and the time you sent it. If you picked a topic, its name. If you arrived from an episode or a post with a tag on the link, that tag. The form also passes your IP address to the bot check and to the rate limit (both explained below), but it is not stored with your enquiry.
If we keep talking: I may keep a contact record: your name, your company, your email and phone number if you give them, where we are in the conversation, and my own notes and a log of our calls, emails and meetings.
When you sign in: Your email address. A sign-in link is only sent to an address that already has an account, and the answer on screen is the same either way. The link works once, for 15 minutes. The database keeps your email, a scrambled (hashed) copy of the link, when it was made, when it expires and when it was used. Then a session cookie keeps you signed in.
When you work with me: Your account: email, name and role. Your client record: a billing email if it differs from your sign-in one, your billing address and company if you give them, the currency you are billed in, and when we started and ended. Your package or engagement: its kind, size, price and how much of it is used.
What "your sessions" means: For each session: its date and time, its length, whether it happened, was cancelled or was missed, and which invoice it was billed on. When you ask for a session through the portal: the times you propose, any context you add, and my reply. If I write a summary or add a transcript or notes for a session, that text is kept too, private to me unless I choose to share it with you in the portal. Reading links I share with you are kept as well.
When you pay: Stripe handles your card, and I never see or store card details. When you buy a package, the site keeps what Stripe passes back: your email, your name, your business name, your billing address and tax ID if you gave them, and your Stripe customer and checkout references. For each invoice: its number, amount, status, dates, the address it went to, the message sent with it, its Stripe reference and payment page, and my note of how it was paid.

Why, and on what legal basis

Each use has one reason the law accepts. The references are to the GDPR, Article 6.

Replying to your enquiry: Steps you asked for before we agree anything. GDPR 6(1)(b).
Keeping a contact record and a private task per enquiry: My legitimate interest in not losing a conversation you started. GDPR 6(1)(f).
Signing you in and running the portal: Our contract. GDPR 6(1)(b).
Sessions, packages and invoices: Our contract. GDPR 6(1)(b).
Keeping invoices and billing records: Tax and accounting law. GDPR 6(1)(c).
The bot check and the rate limits: My legitimate interest in keeping the site and its inbox safe from scripts. GDPR 6(1)(f).
Remembering which episode sent you: My legitimate interest in knowing which episodes reach people. GDPR 6(1)(f).

Nothing on this site relies on your consent today. If that ever changes, I will ask first, and you can withdraw it at any time.

What I do not do

No analytics script and no advertising.
No tracking cookie, and no profile of you across sites.
I do not sell, rent or trade your data.
No AI model reads this site's data. The site's code calls no AI or language model service, so nothing you send here is passed to one.
Nothing a client gives me, and nothing a Praximedes learns, is used to train a model.

Your Praximedes stays on your machine

A Praximedes is built on your own computer, from records you export into a folder you control. You choose what goes in, and what you leave out is not read.

Its record is not sent to me and not sent to this site. The database has no place for it: none of the tables behind the fields above holds it. What I keep about our work is only what the section above lists.

The emails the site sends

A sign-in link, when you ask for one.
A portal invitation, written by me, when your portal is ready.
An invoice, with the PDF attached, and a link to pay it on Stripe when it can be paid online.
A note to me when you write through the form, with your enquiry in it, so that my reply goes straight to you.

The site sends no newsletter and no marketing email. Techne, the podcast and newsletter, lives on Substack. If you subscribe there, Substack holds your subscription as its own controller, under its own privacy policy.

The bot check

The enquiry form can use Cloudflare Turnstile to tell a person from a script. When it is on, it loads only on the page with the form. Cloudflare looks at signals from your browser to make that call, and the site sends Cloudflare the check's token and your IP address to confirm it. The site keeps only the answer: pass or fail. Cloudflare processes this under its own policy.

The form, the sign-in and the link check are also rate limited. To count requests, the site stores a scrambled (SHA-256) value made from your network address or your email, never the address itself, with a count for each hour.

Cookies and browser storage

One cookie, and only after you sign in. It is called praximedes_session and it keeps you signed in. It holds your account number and email address, signed so that nobody can forge it. Page scripts cannot read it, it travels only over a secure connection, and it is not sent when another site links here in the background. Signing out removes it.

If you arrive from an episode with a tag on the link, the browser remembers that tag in its session storage until you close the tab, and then forgets it. It is sent only if you send an enquiry.

Logs

The site's own logs record errors and events, such as a failed email or a payment notice from Stripe. The code keeps email addresses out of them on purpose, and it does not write IP addresses to them.

Cloudflare, which carries every request, sees your IP address and browser details as part of serving the page, and keeps its own records of that under its own retention rules.

Who else handles it, and where

These companies process data for me. Each one has its own privacy policy, linked.

Cloudflare: Hosts the site, runs its server code and the enquiry poller, and runs the bot check on the enquiry form. Where: its global network, so a request is handled near where you are. Policy.
Neon: Hosts the database that holds everything listed on this page. Where: the United States (Amazon Web Services, us-east-1). Policy.
Resend: Sends the site's emails: sign-in links, portal invitations, invoices, and the note that tells me you wrote. Where: outside the European Union. Policy.
Stripe: Takes payments and hosts the pages where you pay a package or an invoice. Where: outside the European Union. Policy.
GitHub: Holds a copy of each enquiry as a private task, so none is missed: your name, the topic, what you wrote and which episode sent you, never your email address. Where: the United States. Policy.

Stripe also uses payment data for its own purposes, such as preventing fraud, as a controller under its own policy.

So your data does leave the European Union. Those transfers rely on the safeguards each provider offers for them, such as the European Commission's Standard Contractual Clauses. Ask me and I will tell you which applies to which provider.

How long

A sign-in link: Works for 15 minutes, and only once. The record that it was sent stays in the database, with the link stored only in scrambled form.
The session cookie: Up to 30 days for a client, or until you sign out.
Rate-limit counts: Each counts for one hour. Counts older than a day are deleted as the site tidies up.
The episode tag in your browser: Until you close the tab.
An enquiry and a contact record: Nothing deletes them on a timer. They are kept while we are talking, and deleted when you ask.
Your account, sessions and portal: For as long as we work together, and then until you ask me to delete them, except what the law makes me keep.
Invoices and billing records: For as long as tax and accounting law requires.

After a deletion, the database's restore history can hold the old data for up to six hours before it is gone.

How it is protected

The whole site is served over HTTPS, and browsers are told to use nothing else.
A sign-in link works once, for 15 minutes, and is stored only as a SHA-256 hash, so a copy of the database signs nobody in.
The session cookie is signed, hidden from page scripts, sent only over HTTPS, and not sent on cross-site requests.
The sign-in page gives the same answer for every address, so it cannot be used to find out who is a client.
Rate limits on the form and the sign-in, counted without storing your IP address or email.
A strict content security policy, so the pages run only the site's own scripts and Cloudflare's.
Your email address is never copied into the private task that tracks an enquiry, and never written to the site's logs.
Card details never reach the site. Stripe holds them.

Your rights

Under the GDPR, you can ask me to do any of these:

Confirm whether I hold data about you.
Show you what I hold, and give you a copy.
Correct it if it is wrong or incomplete.
Delete it, where the law does not make me keep it.
Restrict how I use it while a question is open.
Give it to you, or to someone you name, in a common machine-readable format.
Stop using it on the basis of my legitimate interest (objection).
Anonymise or block data that is not needed or not handled lawfully.
Tell you who I have shared it with.
Withdraw consent, if I ever rely on it, without affecting what was done before.
Not be subject to a decision made only by automated means that affects you significantly, and have such a decision reviewed.

To use any of them, write to hello@praximedes.com from the address I know you by, or tell me how else to confirm it is you. It costs nothing. I answer within one month.

You can also complain to a data protection authority. In Portugal that is the CNPD (cnpd.pt). If you live elsewhere in the European Union, you can go to the authority where you live.

Automated decisions

None. Nothing on this site scores you, profiles you or decides anything about you on its own. The bot check and the rate limits can refuse a single request. If that happens, write to hello@praximedes.com instead.

Children

This site and the work are for adults. They are not meant for children. If you think a child has written to me, tell me and I will delete it.

Changes to this page

When this page changes, the date at the top changes with it. If a change affects how I use a client's data, I will tell clients by email before it applies.

Legal details

The data controller is LVRR, Rua São Paulo, 818, sala 502, Belo Horizonte, MG, 30170-131, Brazil. Where Brazil's LGPD also applies, the matching legal bases are its Article 7, V (contract), II (legal obligation) and IX (legitimate interest), the same rights apply, and you can also complain to the ANPD (gov.br/anpd).