I am Yiğit Demirağ, a PhD student at Institute of Neuroinformatics (INI) of ETH Zurich and University of Zurich, working with Prof. Giacomo Indiveri. I received my MSc from Nanotechnology Research Center (NANOTAM) of Bilkent University, where I worked with Prof. Ekmel Özbay on phase change material based non-volatile memory devices. Previously, I was at Samsung Electronics, EPFL, Google and CERN.

My research focus is to accelerate our path to AGI, by building the perfect accelerator chip for large neural networks. I design nanoscale memory-devices using phase-change materials to mimick the synapses of the biological brain, and study multilayer un/supervised learning on neuromorphic chips. On a side for fun, I experiment with Deep Reinforcement Learning. I usually code in Python (PyTorch and JAX), C++ and MATLAB.