ABOUT
Real-time adaptive AI, built in the Netherlands
Lazy Dynamics builds small, fast audio models that run on-device. Speech enhancement, source separation, voice synthesis — designed for low power, low latency, and real-time use.
Our story
Lazy Dynamics started as a research spin-off from the Signal Processing Systems group at Eindhoven University of Technology. We saw that most audio AI was either too large to run in real time, or too dependent on cloud infrastructure to be practical at the edge.
So we set out to build something different: audio models small enough for embedded hardware, fast enough for live processing, and good enough that you don't need a datacenter to run them.
Today our team is based in the Netherlands and Sweden. We build tools for speech enhancement, source separation and spatial audio simulation.
TEAM
The people behind Lazy Dynamics
Albert Podusenko
CEO
İsmail Şenöz
CTO
Bart van Erp
Chief product officer
Alexey Sidorov
Chief revenue officer
Wouter Nuijten
Senior machine learning engineer
Antoine Buttier
Graduate intern
ADVISORY BOARD
Guided by world-class researchers
Bert de Vries
Co-founder & advisor
Full Professor at the Signal Processing Systems group at TU Eindhoven. Focuses on probabilistic inference and autonomous agents. Creator of the RxInfer framework.
Kevin Murphy
Research advisor
Renowned machine learning researcher and former Associate Professor at the University of British Columbia. Author of influential textbooks on probabilistic machine learning.
PARTNERS & BACKERS
Supported by
The Creator Fund
SSE Business Lab