Software Engineer - High-Performance Computing

This role focuses on improving and evolving software developed in the EP-SFT group such as Geant4 and ROOT, core software frameworks used for particle physics simulation and data analysis. The position involves optimizing and redesigning key components for modern high-performance computing systems, emphasizing GPU acceleration and large-scale parallel processing, supporting future research initiatives like the High-Luminosity LHC and the Future Circular Collider.

Responsibilities

• Define and execute the performance roadmap for Geant4 and ROOT, focusing on GPU architectures (CUDA, SYCL) and advanced programming language features such as C++.
• Implement and refactor critical hot loops, memory layouts, and kernel launches using state-of-the-art tools to identify and eliminate bottlenecks.
• Participate in the organisation of data simulation, processing, and analysis activities within the EP-SFT group.
• Plan projects, organise workshops and trainings, and represent the projects and their vision.
• Engage with user communities, foster collaboration within the SFT group, EP department, CERN, and encourage external contributions from laboratories, universities, and partners.
• Supervise team members and coordinate activities related to EP-SFT data simulation, processing, and analysis projects.

Requirements

• Master's degree or PhD or equivalent relevant experience in the field of Computer Science, Physics or a related field.
• Demonstrated experience in high-performance computing, including GPU programming and C++ (template metaprogramming, SIMD, cache-aware design).
• A proven track record of optimising large (100k+ line) scientific or data-intensive software projects.
• Solid experience with performance profiling tools (e.g., Nsight, VTune, perf, Valgrind), debugging tools (GDB, LLDB), and continuous integration performance regression testing.
• Familiarity with Geant4, ROOT, or other high-energy physics simulation and analysis frameworks would be an advantage.
• Experience contributing to LLVM/Clang or compiler-level optimisation work is considered an asset.
• Knowledge of Monte Carlo particle transport algorithms or statistical data analysis is advantageous.
• A publication record or invited talks at major conferences (e.g., Supercomputing, CHEP, IEEE/ACM performance workshops) is considered a plus.
• Experience in supervising and coordinating external contributors would be desirable.
• Spoken and written English, with a commitment to learn French.

Skills

• High Performance Computing
• GPU Programming
• C/C++ Programming
• Template metaprogramming
• SIMD programming
• Cache-aware design
• Scientific software optimization
• Performance Profiling
• Nsight
• VTune
• perf
• Valgrind
• Debugging Tools
• GDB
• LLDB
• Continuous Integration
• Performance regression testing
• Geant4
• ROOT
• High-energy physics simulation
• LLVM/Clang
• Compiler optimization
• Monte Carlo algorithms
• Statistical Analysis

Languages

English, French