Andrei-Leonard Nicusan
Andrei-Leonard Nicușan is a final-year doctoral researcher in the University of Birmingham’s School of Chemical Engineering and CTO of EvoPhase Ltd., an AI in industry spinout. He published featured articles and Scientific Highlights on machine learning-based algorithms, metaprogramming-driven evolutionary optimisation, simulational-experimental calibration and positron imaging, based on which he won the 2024 IChemE Young Engineers Award for Innovation and Sustainability; his open-source frameworks are actively being used in academia and industry, with work in partnership with GlaxoSmithKline winning the 2023 “Best Use of HPC in Industry” award from HPCWire.
Session
AcceleratedKernels.jl is a unified, backend‐agnostic library for high-performance parallel algorithms in Julia. Built on KernelAbstractions.jl, it lets you write “once” and run everywhere—supporting multithreaded CPUs and GPUs (CUDA, ROCm, oneAPI, Metal) from a single codebase. In this talk I explain the design, implementation, and 200-GPU benchmark results of AcceleratedKernels.jl and show how to write portable code that runs on different hardware without modification.