Postdoc Position in NeuroAI: Biologically Grounded Neural Networks for Cognition

Hosted by the Neurophysics department of the Donders Institute at Radboud University, this postdoc builds the core of the EBRAINS-NEUROTECH tool chain for adaptable brain implants: biologically grounded neural network modules for perception and cognition that run on-device and can be adapted by experimenters. Two use cases guide the work, processing for vision implants and implants for cognitive restoration, and the designs are translated to a neuromorphic architecture together with a postdoctoral researcher at CWI. The project involves teams at the University of Amsterdam (Dr Jorge Mejias) and CWI (Prof. Sander Bohte). The contract is for one year with a possible three-year extension.

The application deadline is 1 December 2026.

PhD in a quantitative field such as physics, mathematics, computer science, engineering or computational or theoretical neuroscience; programming in Python and C with deep-learning tools (PyTorch, TensorFlow, JAX); background in neuroscience, preferably mechanisms of cognitive processes, or a strong wish to acquire it.