| layout |
|---|
default |
Head of Software and Data Science at Bio-Me
Mar 2023 - present
Data Science Architect at Novelda
Feb 2019 - Feb 2023
Designing machine learning algorithms applied to pulse-Doppler radar signals.
Data Scientist at Arundo Analytics
Feb 2017 - Feb 2019
Building machine learning and statistical models for industrial big data applications.
- Created an application for parsing process and instrumentation diagrams (P&IDs) based on a combination of image recognition techniques and OCR.
- Developed ML model to predict ship hull fouling based on noon report data, which matches the performance of numerically costly CFD based models, for clients in the maritime sector.
- Deployed an application for real time visualization and KPI calculations of sensor data from a production process for a large manufacturer of sewing thread.
- Successfully identified several causes of production failures at a chemical plant using Bayesian multilevel regression.
- Created proof of concept of a virtual multi-flow meter for offshore oil and gas wells, combining physical insight with machine learning methods.
Postdoc in Biostatistics at NORMENT (CoE), UiO
June 2015 – Feb 2017
- Characterized geno-/phenotype mapping of mental disorders, focusing on polygenic effects.
- Developed statistical methods to identify pleiotropic effects.
- Work involved heavy lifting of big data containing DNA sequences and patient databases, and running analysis on UIO's high-performance computing facilities.
Jan. 2011 – May 2015
Developed simulations and theoretical models for multiphase flow phenomena, and characterized complex patterning behavior involving granular materials.
- One year residence at Prof. Ruben Juanes group, MIT, developing phase field models for simulating surface tension effect.
- One year residence at ÉOST, Univ. de Strasbourg.
- Results presented in 4 peer reviewed papers, and at 6 international conferences.
Sept. 2010 – Dec. 2010
Taught mathematics and physics at high school level.
June - July 2007 & 2008
Improved power consumption models for clients at Nord Pool based on weather forecasts and historical data.
2015
Thesis: Pattern Formation in Granular Fluids.
- Advisors: Eirik Flekkøy, Knut Jørgen Måløy and Renaud Toussiant
2010
Thesis: Overlap Distributions of Random Cantor Sets and their Applications.
- One year residence at SINP, Calcutta, India. Supervised by Prof. Bikas Chakrabarti.
- Two semesters coursework with focus on theoretical physics, at Univ. of California, Berkeley.
2007
- Teaching assistant in calculus and mathematical analysis