The Center for Legal Data Science (CLDS) is an interdisciplinary research hub at the Faculty of Law of the University of Zurich. The core task of the CLDS is the application, critical reflection and promotion of data science methods in national and international legal contexts. The CLDS pursues the following high-level goals:
-
advancing legal knowledge and legal knowledge production through the application of quantitative and computational methods,
-
enhancing the transparency of law and legal procedures through the application of data science methods,
-
facilitating interdisciplinary legal research and teaching involving quantitative, computational and empirical methods, nationally and internationally.
The position is embedded in the strategic faculty project “Zurich Legal LLM”, led by Prof. Tilmann Altwicker and Prof. Yoan Hermstrüwer. The project develops and evaluates a legal Large Language Model infrastructure for research and teaching at the Faculty of Law. It focuses on building a legally robust and technically reliable Retrieval-Augmented Generation (RAG) system for Swiss legal education and research. The project also develops a Swiss law-specific benchmarking and validation framework for legal question-answering, exam-style reasoning and structured responses to student objections. It addresses key questions of legal AI governance, copyright-sensitive use of doctrinal materials, retrieval transparency, evaluation integrity and faculty-owned AI infrastructure.
information about the CLDS can be found at www.clds.uzh.ch.
As a member of the CLDS team, you will contribute to the technical development and evaluation of the Zurich Legal LLM project. Your responsibilities may include:
-
supporting the design, implementation and documentation of a controlled RAG pipeline for Swiss legal materials,
-
preparing, cleaning and structuring legal teaching materials for retrieval-based use,
-
implementing document ingestion, chunking, embedding, retrieval and citation-tracking workflows,
-
testing and comparing open-weights LLMs and selected commercial systems under controlled conditions,
-
implementing guardrails against content leakage, excessive quotation and extraction-style prompting,
-
supporting the development of a benchmark suite for doctrinal Q&A, exam-style legal reasoning and structured objection-response drafts,
-
contributing to automatic evaluation workflows, including faithfulness checks, citation precision, retrieval quality and structure compliance,
-
maintaining technical documentation, experiment logs and reproducible code,
-
assisting with the development of an internal teaching-oriented prototype for faculty use, including exam-question generation and response-drafting workflows,
-
collaborating with legal researchers, student assistants and faculty members in an interdisciplinary project environment.
The position combines applied machine learning, natural language processing, legal data science and research software development. A legal background is not required, but interest in legal applications of AI is expected.
We offer varied and interesting work in an inspiring and socially relevant environment. Diversity and inclusion are important to us.
In particular, we offer:
-
a challenging applied research position in a motivated and interdisciplinary team,
-
the opportunity to contribute to a strategic AI project at the Faculty of Law of the University of Zurich,
-
hands-on experience with RAG systems, LLM evaluation, legal NLP and AI governance,
-
close collaboration with legal scholars and data scientists,
-
the possibility to contribute to academic publications and internal faculty prototypes,
-
flexible working hours compatible with Master's studies,
-
the possibility of working partly from home,
-
experience in building AI infrastructure for concrete research and teaching use cases.
The following qualifications are required:-
ongoing Master's studies, or recently completed Master's degree, in computer science, data science, computational linguistics, artificial intelligence, software engineering or a related quantitative field,
-
good programming skills in Python,
-
familiarity with machine learning, natural language processing or information retrieval,
-
interest in Large Language Models, Retrieval-Augmented Generation and applied AI systems,
-
ability to work carefully with structured and unstructured text data,
-
good written and oral knowledge of English,
-
motivation to work in an interdisciplinary environment at the interface of law, data science and AI governance.
The following qualifications are assets but not required:-
experience with RAG frameworks, vector databases, graph RAGs, embedding models or LLM APIs,
-
experience with open-weights LLMs and local or controlled model deployment,
-
familiarity with tools such as LangChain, Hugging Face, Ollama, or similar systems,
-
experience with evaluation of LLM outputs, benchmarking, prompt engineering or hallucination/faithfulness testing,
-
experience with document processing workflows for PDF, DOCX, Markdown or OCR-based pipelines,
-
basic knowledge of German, especially for working with Swiss legal teaching materials,
-
interest in legal data science, empirical legal research or AI governance.
Strong motivation, reliability, attention to detail, clear documentation habits and the ability to work independently on technical tasks are particularly valued.
Applicants are requested to submit their application documents (ideally as one single PDF), including:
-
motivation letter,
-
CV,
-
transcript of records,
-
copies of certificates,
-
if available: code samples, GitHub profile, thesis, project report or other evidence of relevant technical work.
Applications are considered on a rolling basis until the position is filled. Job interviews are planned for June 2026.
