💻About Me
I am a postdoctoral researcher at MIT and the Broad Institute, working on reliable and trustworthy AI systems for healthcare and biomedical data. My research focuses on foundation model evaluation, multimodal and longitudinal health data, clinically grounded AI evaluation, and machine learning under missingness. I have worked on wearable foundation models, memorization and privacy risks in foundation models, interpretable machine learning, and large-scale health prediction systems. I completed my PhD in Computer Science at Chalmers University of Technology and have authored publications at venues including ICML, AISTATS, AAAI, and ML4H, including an ICML 2025 spotlight paper.
- Machine learning for health care
- Interpretable AI
- Predictions under missing values
- Foundation Model Evaluation
- Multimodal Learning
- Causal Inference
Postdoctoral Researcher
Massachusetts Institute of Technology (MIT)
PhD Candidate in Machine Learning
Chalmers University of Technology
M.Sc. Information Engineering and Management
Karlsruhe Institute of Technology (KIT)
Visiting Research Student
MIT - Massachusetts Institute of Technology, Cambridge/USA
Visiting Research Student
PreMeDICaL Inria-Inserm team Montpellier, Montpellier/France
At the Healthy ML Lab, we take inspiration from real-world healthcare challenges to develop machine learning models and theory that improve clinical decision-making. Collaborating closely with clinician networks, hospitals, and healthcare providers, we aim to enhance decision-making, improve patient outcomes, and deepen our understanding of complex medical conditions. Our research spans fairness, privacy, interpretability, and human-AI interaction in healthcare, including studying how machine learning systems affect different patient groups, how biases emerge in medical data and models, and how AI can be designed and evaluated to support equitable and trustworthy clinical care.
I am a postdoctoral researcher at MIT and the Broad Institute, working on reliable and trustworthy AI systems for healthcare and biomedical data. My research focuses on foundation model evaluation, multimodal and longitudinal health data, clinically grounded AI evaluation, and machine learning under missingness. I have worked on wearable foundation models, memorization and privacy risks in foundation models, interpretable machine learning, and large-scale health prediction systems.
Let’s collaborate! 🚀
[September 2026] Became a member of the ELLIS Network
[September 2026] Our paper Who Said LLMs Are Better? The Missing Human Baseline and Judge was accepted to Eval4SD and LUHME, with oral presentations at both workshops.
[August 2026] I joined the organizing team of the NeurIPS 2026 workshop Privacy in the Era of Large Opaque Models.
[July 2026] Our paper Clinically Grounded Privacy Evaluation of Medical LMs was accepted to EMNLP 2026.
[June 2026] We released a preprint on Protecting Patient Privacy in Clinical Foundation Models: Technical and Legal Perspectives.
[May 2026] We released a preprint on Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAM. The work was presented at the AI4Good Workshop at ICML 2026.
[Jun 2026] I am giving a talk at KTH on Causal AI for Healthcare.
[Mai 2026] We released a preprint on Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAM.
[Dec 2025] I am traveling to NeurIPS in San Diego.
[Sep 2025] I started my postdoc at MIT.
[Aug 2025] I graduated from my PhD at Chalmers.
[Jan 2025] I am giving a talk at the Traumabase Annual Meeting on a user study with the clinician network, exploring interpretable ML for missing values.
[Nov 2024] My work on Expert Study on Interpretable Machine Learning Models with Missing Values has been accepted to ML4H as a workshop paper. I co-authored the paper Representing Patient History for Interpretable Policy Modeling, also accepted to ML4H 2024.
[Apr 2024] My paper on MINTY: Rule-based Models that Minimize the Need for Imputing Features with Missing Values has been accepted to AISTATS.
[Jun 2023] My paper on Learning Replacement Variables in Interpretable Rule-based Models has been accepted to the 3rd Workshop on Interpretable Machine Learning in Healthcare (IMLH).
[Nov 2022] My paper on Sharing Pattern Submodels for Prediction with Missing Values has been published in AAAI.
[Feb 2021] I co-authored the paper on Predicting Progression & Cognitive Decline in Amyloid-Positive Patients with Alzheimer’s Disease, published in Alzheimer’s Research & Therapy.