Proceedings of the Raptors Conference

Aims and scope

The Raptors Conference is about software that is already running. The talks it selects come from engineers describing systems they built, operated and had to fix: production incidents and what they cost, systems under load, migrations carried out while users were on them, data platforms, applied machine learning that has to hold up outside a notebook, and the security of all of it.

The Proceedings exists because a broadcast is not a record. A conference talk is watched once and then it is difficult to find, difficult to cite and impossible to quote precisely. The work behind it deserves an address that still resolves in five years. That is the whole purpose of this publication.

The scope is deliberately narrow. This is an engineering proceedings tied to one event, not a general-purpose journal, and it does not accept work outside the subject matter of the conference. A submission is in scope when it reports something the author has done, in a production setting, that another engineer could learn from.

What is published here

Each entry in the Proceedings consists of the title of the talk, the author, an abstract of 100 to 200 words written by the author, a link to the presentation slides as supplied, and the date the paper was accepted. Nothing is published under an author's name that the author has not written and approved.

Full papers are not required. Authors who want to submit a longer text alongside the abstract may do so and it is published with the entry. The abstract is what the conference asks for, because the primary work being recorded is the talk.

How a paper enters the Proceedings

Papers come from the conference programme. Talks are submitted to an open call for papers, which for the 2026 edition closed on 28 August 2026 and received 101 submissions. Every submission was scored against a written rubric and the programme was built from those scores.

A Fellow of Hackathon Raptors with prior editorial review experience is seated to ratify the recorded scores and to review the submitted materials for the 2026 edition. That round runs from 12 to 21 October 2026.

Authors of selected talks are invited to contribute an abstract and their slides. Submission is voluntary, it is free, and declining it has no effect on the talk. An author may withdraw an entry at any time by writing to the editorial address, and it is removed.

Frequency and numbering

The Proceedings is published annually, one volume per edition of the conference. The current volume is Vol. 1 (2026), which is in preparation and publishes in October 2026, following the conference on 21 October 2026. Papers accepted for it are listed on this page and each already has its permanent address.

Volume numbering runs in sequence from Vol. 1 (2026). The title of the publication does not change between volumes.

Identifiers, archiving and access

An ISSN has been requested from the ISSN UK Centre at the British Library. The Centre assigns an ISSN after a first issue is published, so the number will appear on this page when it has been allocated, and not before.

A Crossref membership application has been submitted so that every paper can carry a DOI. No DOI has been issued yet and no date is promised for one. When a prefix exists, a DOI is registered against the permanent address each paper already holds, which is why those addresses are fixed now rather than later.

Every paper is published open access on this site with no fee to the author and no fee to the reader. There is no charge of any kind for publication. Slides are served from this site rather than linked to a third-party drive, so the material does not disappear when a link is revoked.

Copyright

Authors retain copyright in their abstract and their slides. Publication here is a non-exclusive licence to host the material and keep it reachable. Authors are free to publish the same work anywhere else, at any time, without asking.

Corrections and withdrawal

Corrections are made on request from the author and the entry records that it was corrected and when. A withdrawn entry leaves a note at its address saying the paper was withdrawn, so that any citation of it still resolves to an honest answer rather than to nothing.

Write to hello@raptors.dev for a correction, a withdrawal, or any question about an entry.