Douw Marx

Background

About

I conduct research in fault detection of machines using signal processing and machine learning methods. I am currently working on my PhD at KU Leuven University, which I will be completing in October 2025. My research is part of the Marie Skłodowska-Curie MOIRA project, studying heterogeneous industrial data. I often find myself proposing the most unconventional ideas in a room of people whilst still being firmly grounded in a responsible data science and a deep physical understanding of vibration, heat transfer and structural mechanics

Work Experience

Marie Skłodowska-Curie PhD Fellow, KU Leuven
Sep, 2021 - Present
Research on fault detection of rotating machines using unsupervised learning and differentiable signal processing methods. Expected end date: 2025-09-30.
- Published 5 peer-reviewed conference papers
- Proposed and lead 4 Masters research projects
- Completed research secondments at INSA Lyon, France and SafranTech, Paris, France
- Runner-up in PHM 2023 Conference Data Challenge, Salt Lake City, Utah, USA
- Participated in advanced machine learning workshops and specialized training programs
SystemModeler Intern , Wolfram MathCore
May, 2021 - Jul, 20212 months
Develop Virtual labs using Wolfram System Modeler and Mathematica.
Mechanical Engineer and Data Analyst , XRAM Technologies
Jan, 2021 - Sep, 20218 months
Build data-driven measurement models for electrode length prediction in electric-arc furnaces (Gaussian process regression and Bayesian structural models).
- Furnace electrode length prediction
- Chemical process data analysis
- Engineering design
- Fluid flow simulations.
Wolfram Summer School Particpant,
Jun, 2020 - Jul, 202020 days
Towards universal robotics
- Cellular Automata
- Inverse Kinematics
Research Masters Student, University of Pretoria
Jan, 2021 - Sep, 20218 months
Towards a Hybrid Approach for Diagnostics and Prognostics of Planetary Gearboxes
- Particle Filters
- Finite Element Analysis
- Inverse problems
- Test bench design
Lab Teaching Assistant, University of Pretoria
Jan, 2017 - Dec, 20203 years 11 months
- Experimental design
- CNC machining
- Design for manufacturing

Skills

Programming/Mathematical Modelling
Python (Pandas, Sklearn, Scipy etc.)

Advanced regularizations in PyTorch

MLOps (MLflow and GuildAI)

Matlab/Octave

C++

Mathematica

Simulink

Modelica
Tools
Linux

LaTeX

Arduino

Version control with Git

Vim

Emacs

HPC with Slurm
Languages
Afrikaans

English

Education

Mechanical Engineering, Masters (With distinction), University of Pretoria
Jan, 2020 - Dec, 2020
Towards a hybrid approach for diagnostics and prognostics of planetary gearboxes
Mechanical Engineering, Honours (With distinction), University of Pretoria
Jan, 2019 - Dec, 2019
Machine Learning

Numerical Methods

Control Systems

Vibration Monitoring

Mechatronics
Mechanical Engineering, B.Eng (With distinction), University of Pretoria
Jan, 2015 - Dec, 2018

Certificates

AI Alignment Course, Bluedot
Issued on: Feb 03, 2025
Introduction to Deep Learning, Coursera
Issued on: Mar 27, 2021
Fundamentals of Accelerated Data Science, NVIDIA
Issued on: Feb 15, 2025

Publications

Constitutional Sensitivities of Preference Models ,
Published on: undefined
Final project for Bluedot AI Alignment Course - investigating how preference models respond to different constitutional frameworks.
Patent for a novel parallel kinematic planar mechanism: WO2020208551A1 ,
Published on: undefined
A 3DOF mechanism with parallel kinematic advantages and stationary actuators.
Five peer-reviewed conference articles ,
Published on: undefined
Research on fault detection in machines using unsupervised data-driven methods and signal processing. See Google Scholar link above.

Interests

Music
TrumpetSingingGuitarBass
Community Service
Volunteer tutoringEffective Altruism
Woodworking
Hiking