MSc den Blanken

PhD student
Electronic Instrumentation (EI), Department of Microelectronics

Themes: Cognitive sensor nodes and systems

Biography

Hey! My name is Douwe den Blanken and I am PhD student in the Cognitive Sensor Systems and Nodes group. I completed my bachelor's degree in Aerospace Engineering (cum laude) here in Delft, followed by a masters in Embedded Systems (cum laude, honors). Throughout my masters, I focused on deep learning (but also vision and RL), custom machine learning accelerator hardware and high-performance (accelerated) computing. For my thesis, under the supervision of Dr. Charlotte Frenkel, I designed and taped out a chip called 'Chameleon' that can do on-chip few-shot learning over temporal data such as speech.

Continuing this, in 2023, I started my PhD in the Cognitive Sensor Systems and Nodes group of Dr. Frenkel. Currently, my research interests are:

ASIC/ML accelerator design
Edge computing/real-life deployments/tinyML
(Extreme) neural network quantization
Transformers & transformer acceleration
Meta-learning/continual learning/unsupervised learning + synergies with each of these
Local/efficient learning rules
Open-source ML software and hardware (currently have 46 public repositories and counting!)

I am always looking for collaborations, so please reach out if you are interested in working together!

Publications

Chameleon: A Multiplier-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data
Blanken, Douwe den; Frenkel, Charlotte;
IEEE Journal of Solid-State Circuits,
pp. 1-16, 2026. DOI: 10.1109/JSSC.2025.3645640

Keywords: ...
System-on-chip;Accuracy;Training;Metalearning;Computational modeling;Vectors;Testing;Prototypes;Integrated circuit modeling;Edge AI;Continual learning (CL);digital accelerator;few-shot learning (FSL);keyword spotting (KWS);multiplier-free;sequential data;system-on-chip (SoC);temporal convolutional network (TCN).

BibTeX support

Last updated: 15 Dec 2025