Louis Ravn

PhD candidate

Louis Ravn is a PhD Candidate on the Deep Culture project at the University of Amsterdam. In this context, his research focuses on the epistemics, politics, and materialities of data and deep learning techniques. His primary research interest lies in better understanding how contemporary deep learning is increasingly shaped by sociotechnical operations of synthesis, compression, and distillation. To that end, he draws from frameworks across science and technology studies (STS), critical AI studies, critical data studies, and social theory. Before joining Deep Culture, he was a Research Assistant at the University of Copenhagen. His research has been published in Big Data & Society, Internet Policy Review, Surveillance & Society, and Digital Society.

https://www.uva.nl/profiel/r/a/l.ravn/l.ravn.html

reflections

Every Token, Every Time: What AI Coding Agents Actually Cost You

We are having a problem with the new AI agent economy. It seems unsustainable for the amount of processing required, as our LLms are not very efficient machines. We built an interactive explainer that makes the economics viscerally clear in your browser. Try it. The Thing Nobody Talks About Every time an AI agent takes […]

reflections

Palantir – a mock-up

Always wanted to know what the excitement about Palantir is? You can now experience Palantir‑style Gotham and Foundry demos – in the Browser. What You’ll See For the Classroom All content opens inside the same page—just click items in the sidebar and explore. Try it at Palantir Demo

research note

The Occasional Bitter Lesson: AI Engineering and Expert Knowledge in the Digital Humanities

Deep Culture Team — University of Amsterdam / deep-culture.org Abstract Rich Sutton’s “Bitter Lesson” argues that general-purpose methods leveraging computation ultimately outperform approaches built on human domain knowledge. We test this hypothesis in the digital humanities by replicating three published experiments: (1) temporal attribution of UN Security Council resolutions, (2) keyword extraction and document classification […]

events

Deep Culture at STS NL 2026: Exploring the Entanglements of Deep Learning and Culture

On April 15-17, 2026, members of the Deep Culture project participated in the inaugural STS NL Conference at the University of Twente within the track “Deep Learning and Culture”. This panel, organized by Louis Ravn, Anna Schjøtt Hansen, Dasha Simons, Paula Helm, and Tobias Blanke, exploring the dynamic relationship between deep learning technologies and cultural […]

research note

What’s the difference between latent space, vector space, feature space, and embedding space?

By Louis Ravn The terms describing the spaces of deep learning (DL) – including latent space, vector space, feature space, and embedding space – are often used interchangeably. While this rightly betrays a close connection between these concepts, they nonetheless differ in important ways. Hence, this blog posts attempts to explain some of the salient […]

events

New Models podcast: Daniël de Zeeuw & Sal Hagen on the post-discursive web

Daniël de Zeeuw (media studies) and Sal Hagen (Deep Culture) were recently on the New Models podcast to discuss ongoing changes in technology and digital cultures; how we’re moving a ‘post-discursive’ web mediated by vectors and vibes were ‘content is empty’. Listen to a preview of the podcast here. The full episode will be made […]

events

Workshop recap: Boundaries & boundary concepts in/as datafication

On February 19-20, members of the Deep Culture project co-organized a workshop on “Boundaries & boundary concepts in/as datafication” together with Yana Boeva (University of Stuttgart) and Sarah Davies (University of Vienna) in Amsterdam. While boundary concepts, such as “boundary objects”, “boundary-work”, “boundary infrastructures”, or “boundary-drawing”, have long played a key role in science and […]

events

From Vectors to Vibes reading group

From February through May, Sal Hagen (Deep Culture) and Daniël de Zeeuw (UvA) will organize a reading group on the theme From Vectors to Vibes. Content Digital platforms have become vast global infrastructures that constantly listen to the everyday murmurs, moods, and habits of users. Ever more intricate deep learning systems allow platforms like TikTok […]

events

Visit to Internet Archive Europe

On Monday, members of the Deep Culture project visited the Internet Archive Europe for an exchange on how the Internet Archive and its WayBack Machine are employed in research on online archives. Jonathan Gray (King’s College London) has written a lovely blog post about the visit on the Internet Archive Europe website. Read it here.

events

GenAI and Creative Practice Conference

Artistic Panel Deep Cultured helped organise the artistic panel at the GenAI and Creative Practice conference in Amsterdam. The panel offered during the a valuable opportunity to examine the continued prominence of individual artistic “mastery” in the context of GenAI—an idea raised by Baptiste Caramiaux in the keynote of the main conference. Femke Herregraven allowed […]