High-Energy Physics Workflows in Native Julia

Julia is emerging as a powerful high-performance language for scientific computing, combining the speed of compiled languages with the flexibility of dynamic ones. The JuliaHEP initiative is advancing this ecosystem by providing native Julia tools for high-energy physics analysis, simulation, and reconstruction pipelines. We present several recent implementations demonstrating Julia’s performance, readability, and modularity compared to traditional C++ and Python workflows. These include Jet Flavour Tagging using inference with ONNX Runtime within the FCC physics framework, Energy Mover’s Distance (EMD) computation for applications of optimal transport in particle physics and the JetReconstruction.jl library for jet clustering, as an alternative to the C++ native FastJet package. Together, these developments illustrate Julia’s growing maturity for large-scale collider analyses and its potential to unify physics data pipelines under a single high-performance language.