Agents that keep learning after they ship.
I'm an AI researcher finishing a PhD at LMU & TUM, with six years at Intel Labs' Neuromorphic Computing group. The through-line of my work: post-deployment adaptation is the missing capability of edge AI. Most deployed models only run inference. I build systems that keep learning locally, on-device, under real latency, energy, and memory budgets.
Predictive world models for robot manipulation
A compact, self-supervised predictive world model with continual learning across multiple time-scales, evaluated on the LIBERO manipulation benchmark. The question I'm chasing: can an agent keep adapting (at test time, across tasks, over its lifetime) instead of freezing the moment it ships?
CLP — Continually Learning Prototypes
A rehearsal-free continual learning algorithm for robots that meet new objects and environments after deployment. It cuts catastrophic forgetting by about 85% and adapts from a handful of labelled examples by reusing pretrained representations.
On-chip continual learning on Intel Loihi 2
A spiking version of CLP co-designed with neuromorphic silicon for sub-millisecond on-device learning. Against the strongest edge-GPU baseline it reaches 113x lower latency and 6,600x lower energy, while matching replay-based accuracy and staying rehearsal-free.
Why I work on frugal, embodied AI
DraftA position, not a neutral engineer's note: against centralized, extractive, brute-force AI; for distributed, frugal, embodied systems that can run on a chip in a field.
The demo-to-deployment gap
IdeaWhy embodied autonomy keeps stalling between a working demo and a deployed system — and what learning on multiple time scales has to do with it.
The next frontier in embodied AI is not a larger pre-trained model, but learning that happens on many time scales at once — milliseconds to years.
I'm drawn to the brain from every angle: neuroscience, psychology, NeuroAI, and the philosophy of intelligence. The same instinct shows up away from the desk — photography, travel, and treating coffee as a small ongoing experiment. This site is meant to show the whole picture, not just the CV.