Python Job: Digital Biomarker Machine Learning Scientist

Digital Biomarker Machine Learning Scientist

• Build research-grade deep learning models for human activity recognition, anomalous behavior detection and clinical outcome prediction, to apply to highly complex sensor data from smartphones, wearables and other devices in clinical studies.
• Identify relevant recent advances in statistical and/or machine learning methods for longitudinal, high dimensional and sensor data, and adapt them for our analysis needs.
• Participate in the implementation and deployment of production-level deep learning models for our data processing pipelines.
• Support analyses that help establish the biological, clinical and patient relevance of sensor data collected in clinical studies.
• Drive the understanding of the output of machine learning models by clinical development programs through the development of compelling narratives and presentations.
• Contribute to manuscripts for peer-reviewed journal articles to communicate our methodological advances, as well as establish the biological, clinical and patient relevance of data collected in novel digital biomarker approaches.

• PhD or Masters level degree in machine learning, mathematics, statistics, engineering, natural science or informatics
• Min. 3+ years of experience as a machine learning scientist building deep learning solutions in a research environment
• Hands-on experience with common machine learning frameworks such as PyTorch or TensorFlow
• Excellent understanding of common machine learning concepts (datasets splitting build/validation, cross validation, overfitting, etc.)
• Deep understanding of theoretical underpinnings of relevant methods, such deep learning architectures (ConvNets, Transformers, LSTMs), optimization or evaluation
• Facility in Python, R, Matlab or similar software packages/languages
• Highly efficient user of MS Office or Google docs, proficient at slide design
• Excellent English speaking and writing skills
• Excellent analytical and communication skills; ability to communicate complex machine learning concepts in lay terms to very diverse and heterogeneous audiences
• Experience working independently within a collaborative team
• Publication record appropriate for career level

Nice to Haves

• Roche experience
• Experience in a scientific role in the pharmaceutical industry, ideally involving clinical studies
• Previous hands-on clinical experience with neurological diseases such as Multiple Sclerosis, Parkinson's Disease, Autism Spectrum Disorder or Schizophrenia
• Experience with Agile methodologies and ceremonies
• Experience with Bayesian methods or longitudinal data analysis
• Experience with remote patient monitoring
• Experience in experimental/clinical study design
• Experience with MLOps

Application process