I am a Machine Learning Researcher in Sensyne Health with a special emphasis on causal inference. Together with my team, we develop tools to establish causal connections in the millions of Electronic Health Record patient data that this start-up possesses. A central problem across natural science is identifying general laws of cause and effect, and is extremely relevant in the medical domain as well as for policymakers. To do so we use classical tools such as inverse probability weighting, propensity score matching, or dynamic bayesian networks as well as more advanced methods such as causal trees or double machine learning.
Previous to joining Sensyne Health I obtained my Ph.D. in Economics at Glasgow University under the examination of Michael McMahon. In my thesis, I used Natural Language Processing (NLP) techniques to model a wide range of uncertainty/risk indices, from policy uncertainty for different European countries (in their own language), political uncertainty in Scotland, to narratives concerning cryptocurrencies. To do so I used news and social media and NLP such as topic modeling, words embeddings, and sentiment analysis. My codes from published papers can be found in the research section. They are mainly written in Python and to a lesser extent in R. For more information regarding recent projects you can also visit my Github account.
During my Ph.D. I worked in the European Central Bank (ECB) where I carried out research on how uncertainty propagates over time and in which ways it connects to the real economy. I continue to do some work as a consultant at the ECB on regular basis and also continue with research in the field of economics with the Adam Smith Business School from the University of Glasgow. If you or your organisation think you can benefit from my skills, please do not hesitate on contacting me.