Perhaps you agree that it is the task of an engineer to perceive, understand and control his environment. However, reality remains upredictable, noisy and full of unexpected influences. So how can we manipulate a system to the best of our ability? An interesting question with many answers.
My name is Stijn Woestenborghs. I am an engineer at Sony with a MSc in Control Engineering and Automation. I am genuinely passionate about the myriad of advanced technologies that exist today and I get excited to see how they affect the lives of others.
With this site I would like to share with you what I have learned or what inspires me. I want to touch upon the subjects that interest me most and go deep into the implementation that lies at the core of a solution.
After all, something remains just a theory before it gets
actually deployed.
Mojo does give Superpowers
OCTOBER, 2023 topic: Machine Learning
Faster than Python, faster than Numpy, faster than JAX and faster than C++, is my experience when testing out Mojo for the first time. A simple optimization problem implemented in Mojo: A new programming languague for AI developers that is on its journey to become a full superset of Python.
Unlocking the Car Battery's full Potential
APRIL, 2023 topic: Machine Learning
The rising efficiency and energy density of Li-ion batteries is driving an exciting wave of innovation, which not only accelerates adoption, but also enables new technologies with an even greater impact. Needless to say that an accurate State of Charge estimation is vital.
Python, C++ ... Let's Talk!
AUGUST, 2022 topic: Software Engineering
In an environment where complex optimization algorithms are being developed for edge computing platforms with limited resources, choosing the right tech stack is curcial. What about Python vs. C++ or the benifits of both?
Sim2Real transfer in Mechatronic Systems
JULY, 2021 topic: RL - Adaptive Domain Randomization
While the idea behind Reinforcement Learning for mechatronic systems is appealing, it struggles to find its way into an industrial setting. Simreal, a library for Adaptive Domain Randomization.
The Future of Time Series Forecasting
MARCH, 2021 topic: ML - Time Series Forecasting
After reinforcement learning, computer vision & NLP, time series forecasting is next in line to be disrupted by deep learning technology.
A case study on energy demand forecasting
