
Jaeyoon Lee developed machine learning-based candidate selection features for the AliceO2Group/O2Physics repository, focusing on the XicToXiPiPi analysis within the PWGHF framework. He introduced a new C++ class to handle ML feature extraction and inference, enabling flexible model loading from both CCDB and local files. Jaeyoon expanded ML-ready data structures by adding THnSparse axes for prompt and non-prompt BDT scores, supporting more granular data analysis. He also improved Monte Carlo matching and vertexing precision, directly enhancing candidate selection accuracy. His work demonstrated depth in C++ development, data analysis, and high energy physics, resulting in robust, production-ready analysis workflows.
February 2025 performance summary for AliceO2Group/O2Physics. Focused on Xic analysis improvements through ML-ready data expansion and critical fixes to MC matching and vertexing. Demonstrated impact on data quality and ML readiness, with direct benefits to physics candidate selection and downstream analytics.
February 2025 performance summary for AliceO2Group/O2Physics. Focused on Xic analysis improvements through ML-ready data expansion and critical fixes to MC matching and vertexing. Demonstrated impact on data quality and ML readiness, with direct benefits to physics candidate selection and downstream analytics.
Month 2024-11 – Key accomplishments in AliceO2Group/O2Physics (PWGHF): Delivered ML-based XicToXiPiPi candidate selection within PWGHF. Introduced HfMlResponseXicToXiPiPi to perform ML feature extraction and inference and integrated it into the taskXicToXiPiPi analysis task. Enabled loading ML models from CCDB or local files for flexible deployment. Major bugs fixed: none reported for this repository this month. Overall impact: enhances Xic candidate selection quality and workflow efficiency, providing a reusable ML-enabled path in PWGHF and a solid foundation for ML-driven analyses. Technologies/skills demonstrated: ML integration in HEP pipelines, C++/ROOT, CCDB I/O, model loading from multiple sources, software design, and version control traceability (commit fe5032f7b3e4ac568fd92d8e478d5e19c9498c97).
Month 2024-11 – Key accomplishments in AliceO2Group/O2Physics (PWGHF): Delivered ML-based XicToXiPiPi candidate selection within PWGHF. Introduced HfMlResponseXicToXiPiPi to perform ML feature extraction and inference and integrated it into the taskXicToXiPiPi analysis task. Enabled loading ML models from CCDB or local files for flexible deployment. Major bugs fixed: none reported for this repository this month. Overall impact: enhances Xic candidate selection quality and workflow efficiency, providing a reusable ML-enabled path in PWGHF and a solid foundation for ML-driven analyses. Technologies/skills demonstrated: ML integration in HEP pipelines, C++/ROOT, CCDB I/O, model loading from multiple sources, software design, and version control traceability (commit fe5032f7b3e4ac568fd92d8e478d5e19c9498c97).

Overview of all repositories you've contributed to across your timeline