Publisher
source

University of Basel

Just Landed

Postdoctoral Researcher in Language Model Interpretability and Digital Humanities University of Basel in Switzerland

Degree Level

Postdoc

Field of study

Computer Science

Funding

Postdoctoral position at 80% for 48 months within an SNSF-funded project. First contract is for two years, with extension for the remaining two years expected after positive review. Salary follows University of Basel and SNSF guidelines; access to sciCORE and GPU resources is provided. Teaching in the Digital Humanities programme may be arranged.

Deadline

Nov 25, 2026

Country flag

Country

Switzerland

University

University of Basel

Social connections

How do I apply for this?

Sign in for free to reveal details, requirements, and source links.

Apply for this position

Keywords

Computer Science
Information Technology
Philosophy
Natural Language Processing
History
Literature
Digital Humanities
Linguistics
Large Language Models
ML

Suggested positions

About this position

Postdoctoral Researcher (80%) in Language Model Interpretability and Digital Humanities at the University of Basel, within the Digital Humanities Laboratories (DHLabs) and the research group contralab led by Prof. Dr. Ranjodh Singh Dhaliwal. The project, A Philology of Probabilistic Texts: The Problem of Interpretation in Artificial Intelligence, is funded by the Swiss National Science Foundation (SNSF).

The position sits at the intersection of computer science, natural language processing, machine learning, digital humanities, historical texts, literary theory, media studies, STS, and the philosophy of language. The successful candidate will lead the computational work on a four-year project studying how large language models interpret historical texts, how eighteenth-century concepts shift across genres and decades, and how fine-tuned open-weight models represent those concepts.

Key responsibilities include preparing and validating historical corpora, fine-tuning transformer language models, building attribution and interpretability pipelines, running controlled experiments, and leading the technical release of an annotated corpus, benchmark, and leaderboard. The role also includes co-authoring papers for NLP/ML venues and humanities publications.

Applicants should hold a doctorate by the time of appointment in a relevant computational field or digital humanities area, and should have hands-on experience with PyTorch, Hugging Face, GPU clusters, interpretability methods, and reproducible experimental practice. A strong interest in historical texts, literary theory, and the philosophy of language is essential; German is welcome.

This is a postdoctoral opening, 80% FTE, for 48 months. The first contract is for two years, with extension expected after positive review. Salary follows University of Basel and SNSF guidelines, and the project provides access to sciCORE and GPU resources. Residence in the trinational region of Basel is mandatory. International applicants are welcome, and relocation/permit support is available.

Application deadline: 25 November 2026.

How to apply: Submit via the online portal with an academic CV and a single PDF containing a cover letter and one research example. You may also provide the full URL of a code repository, dataset, or digital project as plain text. Questions can be sent to [email protected].

Funding details

Postdoctoral position at 80% for 48 months within an SNSF-funded project. First contract is for two years, with extension for the remaining two years expected after positive review. Salary follows University of Basel and SNSF guidelines; access to sciCORE and GPU resources is provided. Teaching in the Digital Humanities programme may be arranged.

What's required

A doctorate by the time of appointment in computer science, NLP or computational linguistics, machine learning, digital or computational humanities, or a related field with a substantial computational component. Hands-on experience training, fine-tuning, and analysing transformer language models in Python (PyTorch, Hugging Face), including on GPU clusters. Experience with interpretability methods such as feature attribution, probing, activation patching, sparse autoencoders, and/or building evaluation datasets and benchmarks. Experience designing controlled experiments with baselines, statistical care, and reproducible research-software practice. Strong interest in historical texts, literary theory, and the philosophy of language. Strong English; German is welcome. Desirable: experience with historical or OCR-derived text, semantic change detection, annotation studies, released datasets or open-source tools, publications in NLP or ML, and engagement with literary theory, historical materialism, hermeneutics, or structuralist/historical linguistics.

How to apply

Apply via the online portal by 25 November 2026. Submit an academic CV as a separate file and one PDF containing a cover letter and one research example; alternatively, provide the full URL of a code repository, dataset, or digital project as plain text. State in the cover letter whether you currently hold or will hold another job alongside this position.

More information can be found here

Ask ApplyKite AI

Start chatting
Can you summarize this position?
What qualifications are required for this position?
How should I prepare my application?