AlgoPrivacy Policy
Last updated 16 September 2026
The short version. AlgoStudio sets no cookies, has no accounts, and loads no advertising or cross-site tracking scripts. It stores the recordings of algorithm runs for fourteen days and then deletes them. Nothing you do here is sold, shared or profiled.
Who runs this
AlgoStudio is a final-year engineering project, operated by its author as a free educational demonstration. It is not a company and not a commercial service. Questions about this policy go to the project's issue tracker, which is public.
Cookies
This site sets no cookies — not for analytics, not for preferences, not for anything. That is why you have not been shown a cookie banner: there is nothing to consent to, and a banner asking permission for cookies that do not exist would be theatre.
The site does use your browser's local storage for one
purpose: remembering the sizes of the panels you have dragged, under the
keys algostudio.layout.v1.rows and
algostudio.layout.v1.cols. This data is written by your
browser, stays on your device, is never transmitted to the server, and
disappears when you clear site data.
What the server stores
| What | Why | Kept for |
|---|---|---|
| An execution recording: which algorithm was run, its inputs, the resulting event stream, timings and status | It is the thing you are stepping through. Re-used if the same algorithm is run again, so a repeated demo costs one execution. | 14 days, then deleted automatically |
| A shared session: a random token plus the interface state, created only if you explicitly share a run | So a link you send reopens at the same step | Deleted with its recording |
| Your IP address, held in memory only | Rate limiting, so one visitor cannot exhaust a free server | 60 seconds, in memory, never written to a database |
There is no account system, so none of the above is linked to an identity. No name, email address, or device fingerprint is collected or derived.
Code you write
Code you write in the editor never leaves your browser. It is executed on your own machine by a Python runtime the page downloads once, so it is never uploaded, never stored here, and never seen by this server — which continues to refuse to execute source it did not ship with. There is nothing for us to keep, because nothing arrives.
The consequence worth knowing: a run of your own code exists only in that tab. Closing it discards the recording, and there is no link that would let anyone else open it.
Server logs and hosting
This site is hosted on a third-party platform. Like every web server, it produces request logs that include IP addresses, timestamps, requested paths and user-agent strings. These are generated by the hosting platform and the web server, retained on the platform's own schedule, and used only for diagnosing faults and abuse. They are not analysed, combined with other data, or used to build any profile of you.
Analytics
This deployment runs no analytics by default, and none is enabled unless the operator explicitly configures one. If analytics is ever switched on, it will be a cookieless aggregate counter — page-level counts with no cookies, no cross-site identifiers and no personal data — and this section will say which provider and link to their policy before it is enabled. Third-party advertising and behavioural tracking will not be added.
If a provider that stores anything on your device were ever used, the site would ask first: the script would not load until you accepted, declining would be one click and would be remembered, and a Cookie settings control would let you change your mind. That machinery is built and switched off, because nothing currently needs it.
The AI tutor
The explanation panel works in two ways. Its default is a template explainer that is built from the recorded event stream and runs entirely on this server, sending nothing anywhere. If an operator supplies an API key, questions and the relevant slice of the execution are sent to that language model provider to compose an answer.
On this deployment, no API key is configured, so the template explainer is what you get and no request leaves this server.
What is never done
- Your data is not sold, rented, or shared with advertisers.
- No behavioural profile, audience segment or fingerprint is built.
- There are no third-party social, advertising, or session-replay scripts.
- Nothing you do here is used to train a machine-learning model.
Your choices
Because there are no accounts, the controls are direct ones:
- Delete a recording now — recordings are removed automatically after fourteen days, and you can ask for a specific one to be deleted sooner by opening an issue with its execution id.
- Clear local preferences — clearing site data in your browser removes the stored panel sizes. Nothing on the server refers to them.
- Run it yourself — the whole project is open source. Run it on your own machine and no data reaches anyone else at all.
If you are in a jurisdiction granting rights of access, correction or erasure, they apply to the limited data described above; ask through the issue tracker and include the execution id concerned.
Children
The site is an educational programming tool suitable for general audiences. It does not knowingly collect personal information from anyone, of any age, because it does not collect personal information at all.
Changes to this policy
If what is stored changes, this page changes with it and the date at the top is updated. Because the site is open source, the full history of this page is visible in the repository.