PhD position in Developing Language-Model Methods for Pediatric Critical Care

Kinderspital Zürich - Eleonorenstiftung; Zürich; Vor Ort; CHF 120.000 - 150.000; Vollzeit

Zusammenfassung

Kinderspital Zürich – Eleonorenstiftung in Zürich, Switzerland, invites applications for a Senior Scientist to develop trustworthy language-model methods for pediatric critical care and to lead rigorous evaluation across multiple facets of model performance.

You will collaborate with clinicians and data scientists to translate clinical needs into robust studies and publish findings in premier ML, NLP, and medical-informatics venues, contributing reproducible research code.

Qualifikationen

• Master's degree in computer science, data science, biomedical engineering, computational linguistics, statistics, or closely related field.
• Strong foundation in machine learning and hands-on experience with deep learning; experience with transformers, LLMs, or retrieval systems desirable.
• Proficiency in Python and PyTorch or similar frameworks.
• Experience with cluster environments (e.g., SLURM).
• Interest in methodological research with clinically grounded validation.
• Clear communication and ability to collaborate across technical and clinical disciplines.
• Strong English communication; German knowledge advantageous.

Aufgaben

• Develop trustworthy and efficient language-model methods for pediatric critical care.
• Perform rigorous evaluation protocols covering accuracy, calibration, faithfulness, fairness, efficiency, and clinical utility.
• Work closely with clinicians, data scientists, and research partners to translate clinical needs into technically sound studies.
• Publish in leading machine-learning, NLP, and medical-informatics venues and contribute reproducible research codes.

Kenntnisse

Machine learning

Deep learning

Transformers

LLMs

Retrieval systems

Python

PyTorch

Cluster environments

SLURM

Communication

Collaboration

English

German

Ausbildung

Master's degree in computer science, data science, biomedical engineering, computational linguistics, statistics, or related field

Jobbeschreibung

Your responsibilities

• Develop trustworthy and efficient language-model methods for pediatric critical care.
• Perform rigorous evaluation protocols covering accuracy, calibration, faithfulness, fairness, efficiency, and clinical utility.
• Work closely with clinicians, data scientists, and research partners to translate clinical needs into technically sound studies.
• Publish in leading machine-learning, NLP, and medical-informatics venues and contribute reproducible research codes.

Your profile

• A Master's degree in computer science, data science, biomedical engineering, computational linguistics, statistics, or closely related field
• Strong foundation in machine learning and hands-on experience with deep learning; experience with transformers, LLMs, or retrieval systems is highly desirable
• Proficiency in Python and a modern deep-learning frameworks such as PyTorch
• Experience with cluster environments (e.g., SLURM)
• An interest in methodological research as well as careful, clinically grounded validation
• Ability to communicate clearly and collaborate across technical and clinical disciplines
• Strong communication skills in English, Knowledge of German is an advantage

Daphné Chopard
Senior Scientist
[phone]

Kinderspital Zürich as an employer

Kinderspital Zürich is the largest university children's hospital in Switzerland with about 2,500 employees, and is one of the leading centres for paediatric and youth medicine in Europe. Although the hospital may be big, it's the little ones that count: up to 100,000 patients are cared for each year across 47 faculties, where we do everything in our power to ensure their well being. 'Kispi' not only offers cutting-edge devices and the highest quality of care, but also a good sense of interpersonal issues. Alongside its comprehensive range of paediatric medicine and surgery facilities, the children's hospital also offers a unique rehabilitation centre and a research centre.