In the applied mathematics department, we explore novel ways of accelerating inference in convolutional neural networks on FPGAs.
SPiKE.AI is a science hub, comprising of computer and data specialists, engineers and neuroscientists that capitalizes on interdisciplinary thinking to hack innovation at the interface of machine learning, intelligent systems and system design in a wide range of private and public services.
I was part of a large project investigating the impact of mutations on synapses in mammalian CNS. My role consisted in formulating novel hypotheses based on models I developed during my PhD and testing them in an experimental setting.
This position was an opportunity to continue working on projects started during my PhD. Notably: comparative study of metabolic cost and reliability of action potential transmission in myelinated and unmyelinated axons using stochastic simulation. - Impact of channel noise on the waveform of action potentials and EPSPs - Homeostatic constraints on the function of neural fibres
My research focused on biophysical constraints faced by neural fibres (energy, noise, time and volume), and how evolution of nervous systems reflects trade-offs between those constraints. I maintained and developed libraries for stochastic simulation of ion channels and neural fibres (Modigliani, available from modigliani.co.uk) and used them to simulate models that I created based on experimental data.
Client infrastructure auditing and analysis. Project planning, elaboration and implementation of IT security strategies for large French companies.
I specialised in theoretical computer science, including a course on non-classical programming paradigms (Erlang, Haskell, Lisp, etc.) I successfully completed the embedded systems module (ROSE), and followed Dr. Dessalles's courses on artificial intelligence and cognitive sciences.
Development of a network performance monitoring application used by financial institutions. I studied different proprietary network communication protocols and made a survey of available timestamp solutions.
Development of additional functionalities for online bug tracking and task management tools.
Master level class, department of Computing
3rd year class in the department of Bioengineering
Master level class in the department of Bioengineering