11/12/2026 updated
HM
60 % partially available
Healthcare AI / Machine Learning Engineer | Clinical Time Series & EHR Predictive Modeling
Aalborg, Denmark
Worldwide
PhD in Computer Science (expected 2026), Aalborg University, DenmarkHealthcare AI & Clinical Data Modeling
Expertise in EHR/ICU data processing including eICU-CRD and MIMIC-III, clinical forecasting, imputation, informative missingness modeling, clinical risk prediction, and privacy-constrained hospital data analysis using the OMOP Common Data Model.
Machine Learning & Deep Learning
Proficiency with PyTorch, TensorFlow/Keras, scikit-learn, Transformers, Prior-Data Fitted Networks (PFNs), diffusion-based generative modeling, self-supervised learning, and XGBoost for building and benchmarking advanced ML models.
Time Series Analysis & Predictive Modeling
Strong background in multivariate and irregular time series forecasting, missing-data modeling, representation learning, domain shift, cross-domain evaluation, model benchmarking, and ablation studies for complex clinical datasets.
Programming Languages
Solid command of Python, SQL, MATLAB, NumPy, pandas, and SciPy for data preprocessing, feature engineering, and model development.
Engineering & DevOps Tools
Hands-on experience with Docker, Git, Linux, PostgreSQL, Django, ETL/data pipelines, Streamlit, relational data modeling, logging, and caching.
Foundation Models & Generative Modeling
Design and implementation of foundation-model approaches combining diffusion-based generative modeling with Prior-Data Fitted Networks for clinical time series imputation and forecasting.
Biomedical Signal Processing
Development of EEG signal-processing and ML methods for brain-computer interfaces, including cross-subject transfer approaches and spectral-hashing methods for medical image retrieval.
Data Pipeline & ETL Development
Building relational schemas and ETL pipelines using Python, Django, pandas, and PostgreSQL, including transformation logic, logging, caching, and database maintenance.
Expertise in EHR/ICU data processing including eICU-CRD and MIMIC-III, clinical forecasting, imputation, informative missingness modeling, clinical risk prediction, and privacy-constrained hospital data analysis using the OMOP Common Data Model.
Machine Learning & Deep Learning
Proficiency with PyTorch, TensorFlow/Keras, scikit-learn, Transformers, Prior-Data Fitted Networks (PFNs), diffusion-based generative modeling, self-supervised learning, and XGBoost for building and benchmarking advanced ML models.
Time Series Analysis & Predictive Modeling
Strong background in multivariate and irregular time series forecasting, missing-data modeling, representation learning, domain shift, cross-domain evaluation, model benchmarking, and ablation studies for complex clinical datasets.
Programming Languages
Solid command of Python, SQL, MATLAB, NumPy, pandas, and SciPy for data preprocessing, feature engineering, and model development.
Engineering & DevOps Tools
Hands-on experience with Docker, Git, Linux, PostgreSQL, Django, ETL/data pipelines, Streamlit, relational data modeling, logging, and caching.
Foundation Models & Generative Modeling
Design and implementation of foundation-model approaches combining diffusion-based generative modeling with Prior-Data Fitted Networks for clinical time series imputation and forecasting.
Biomedical Signal Processing
Development of EEG signal-processing and ML methods for brain-computer interfaces, including cross-subject transfer approaches and spectral-hashing methods for medical image retrieval.
Data Pipeline & ETL Development
Building relational schemas and ETL pipelines using Python, Django, pandas, and PostgreSQL, including transformation logic, logging, caching, and database maintenance.
Experience
Professional experience5 - 10 years
Freelancer since2020
Languages
EnglishFluent
Project history
Developing ML methods for sparse, irregular, and heterogeneous electronic health records with applications to ICU forecasting, imputation, informative-missingness modeling, and clinical prediction. Building end-to-end experimental pipelines and Transformer-based approaches for ICU blood-glucose prediction.
Developed missingness-aware foundation models for clinical time-series imputation and forecasting by combining generative modeling with Prior-Data Fitted Networks. Worked with privacy-constrained real-world hospital data and OMOP-based clinical data standardization within governed analytical environments.
Built recommendation algorithms and supporting data workflows; designed a PostgreSQL schema and implemented Python/Django/pandas ETL pipelines with logging and caching.