Home Institution: Princeton University

Ianna Osborne

Research Software Engineer, Princeton University
Contact me:

My research:

I am a Research Software Engineer specializing in open-source software development within the Python ecosystem and Julia for fundamental physics. My work focuses on designing and optimizing software for high-energy physics (HEP) data analysis.

In the past, I contributed extensively to CMS software (CMSSW), particularly in geometry description, simulation, and event display. My expertise spans software architecture, numerical optimization, and large-scale scientific computing, with a strong focus on enhancing data analysis tools for physics research.

My expertise is:

  • Programming Languages: C++, Python, Julia
  • Technologies: Awkward Array, Numba, Just-in-Time Compilation
  • Integration: Python integration with other languages
  • Software Development: Open-source contributions in Python and Julia ecosystems
  • Scientific Computing: Geometry modeling, simulation, event visualization
  • Configuration & Deployment: Managing and distributing complex software environments

A problem I’m grappling with:

I am currently focused on integrating Awkward Array with Numba to optimize performance through Just-in-Time compilation techniques. Additionally, I am exploring Python on GPU to leverage parallel computing power for complex data analysis.

I’ve got my eyes on:

  • High-performance computing (HPC) applications in fundamental physics
  • Large-scale parallelism and deep storage hierarchies for HEP data
  • Julia’s potential for optimizing scientific computing workflows

I want to know more about:

To further improve my work, I am keen to deepen my understanding of:

  • Optimizing data movement and storage for large-scale scientific workflows
  • Heterogeneous computing to leverage CPUs, GPUs, and specialized accelerators
  • Efficient memory management strategies for structured and unstructured data
  • Scalable distributed systems that handle massive physics datasets

By exploring these areas, I aim to contribute to more efficient and scalable computational solutions for HEP.