2023-2025 Cohort

Matt Kwiecien

Matt Kwiecien is a third-year student in the Physics Department at the University of California, Santa Cruz. He is part of the Legacy Survey of Space and Time (LSST) Dark Energy Science Collaboration (DESC). Matt will be working with university mentor Tesla Jeltema (UCSC) and lab mentor Marc Paterno (Fermilab) in the area of collaborative software infrastructure, specifically on development of the Firecrown Pipeline for LSST DESC. He will train in the development of scientific pipelines and software architecture for experiments. Matt entered graduate school at UCSC after working in industry as a software engineering team lead.

Johannes Wagner

Johannes Wagner is a fourth-year Physics Ph.D. student at the University of California, Berkeley. He is passionate about designing algorithms for high-energy collider physics. Johannes will focus on collaborative software infrastructure in heterogeneous computing and statistical data analysis. He plans to leverage compute-heavy machine learning for jet flavor tagging with the ATLAS detector, in collaboration with his university mentor Heather Gray (UCB) and lab mentor Simone Pagan Griso (LBNL). As an undergraduate, Johannes minored in computer science and data science to build a solid foundation in computing.

Eric Le

Eric Le is a second-year Ph.D. student in Physics at the University of California, Irvine. He works with the ATLAS group at UCI. Eric is focused on the area of hardware-software co-design in the integration of GPUs and FPGAs for triggers and track reconstruction. He will work with university mentor Anyes Taffard (UCI) and lab mentor Viviana Cavaliere (BNL) on optimizing algorithms and improving host-kernel data transfer for event filter tracking. Eric will also develop generic host code to load kernels across FPGAs and GPUs and maintain data transfer queues across different architectures.

Luc Le Pottier

Luc Le Pottier is a third-year student in the Physics Ph.D. program at the University of California, Berkeley. He works in the area of hardware-software co-design on FPGA-based data acquisition systems for high-energy experiments. Luc will work with university mentor Haichen Wang (UCB) and lab mentor Timon Heim (LBNL) on the development of a high-speed readout system for a pixel detector beam telescope. He is interested in improving the throughput of the system with careful attention to the hardware-software interface and limitations. In the past, Luc has lectured on machine learning in physics for undergraduates.

Russell Marroquin Solares

Russell Marroquin Solares is a second-year Physics Ph.D. student at the University of California, San Diego. His interests are in hardware-software co-design, specifically the implementation of advanced computational algorithms for machine learning and triggering technologies. Russell will collaborate with university mentor Javier Duarte (UCSD) and lab mentor Nhan Tran (Fermilab) to enhance the hls4ml tool for new FPGA firmware, making it possible to apply machine learning principles in FPGAs for the CMS experiment. He will also investigate pruning and supporting sparse ML models in hls4ml.

2024-2026 Cohort

Jade Chismar

Jade Chismar is a first-year Physics Ph.D. student at the University of California, San Diego. Her interests are in the area of high-performance computing, specifically particle tracking algorithms for high-energy physics. Jade will work with university mentor Frank Wuerthwein (UCSD) and lab mentor Giuseppe Cerati (FNAL) on a line-segment tracking (LST) algorithm for the CMS experiment. This algorithm is designed to be highly parallelizable and vectorizable on modern processors. Jade will investigate the machine learning possibilities for LST.

Miles Cochran-Branson

Miles Cochran-Branson is a first-year Physics Ph.D. student at the University of Washington. His interests are related to hardware-software co-design, especially in machine learning as-a-service with heterogeneous architectures. Miles will collaborate with university mentor Shih-Chieh Hsu (UW) and lab mentor Xiangyang Ju (LBNL) to investigate GPU algorithms for particle tracking in the ATLAS experiment. This project will build on the ACTS tracking library and the Exa.TrkX software. Miles will develop a working model for the use of ACTS-aaS and Exa.TrkX-aaS at the LHC.

Alex Golub

Alex Golub is a third-year Physics Ph.D. student at the University of Washington. They work with the ATLAS group at UW. Alex’s interests are in machine learning and collaborative software infrastructure. Alex will work with university mentor Gordon Watts (UW) and lab mentor Ben Nachman (LBNL) on developing algorithms to differentiate between charged and neutral particles in the ATLAS calorimeter. This is part of the Global Particle Flow reconstruction software, but it can eventually be applied to other detectors, too. Alex has previously used recurrent neural networks to identify calorimeter signatures of long-lived particles.

Luke Grossman

Luke Grossman is a second-year student in the Physics Ph.D. program at the University of California, Berkeley. His interests are in the area of collaborative software infrastructure, specifically the data processing flow and enhanced tracking algorithms. Luke will collaborate with university mentor Marjorie Shapiro (UCB) and lab mentor Simone Pagan Griso (LBNL) on extended tracking performance using the ATLAS EventIndex catalog. One goal is to make the late processing of selected data more universally applicable in the collaboration.

Charlie Hultquist

Charlie Hultquist is a second-year Physics Ph.D. student at the University of California, Berkeley. His interests lie in high-performance computing, most directly in foundational models for machine learning. Charlie will work with university mentor Haichen Wang (UCB) and lab mentors Steven Farrell and Xiangyang Ju (LBNL) on a prototype system featuring multi-task pretraining and transfer learning. This work leverages the Perlmutter high-performance supercomputer at LBNL, requiring special data handling and resource optimization.

Hadley Santana Queiroz

Hadley Santana Queiroz is in the second year of the Physics Ph.D. program at the University of California, Berkeley. She works in the area of high-performance computing and machine learning, with some connections to collaborative software infrastructurein ATLAS. Hadley will collaborate with university mentor Heather Gray (UCB) and lab mentor Paolo Calafiura (LBNL) on data augmentation for graph neural networks. The data augmentation methods are implemented in the common GNN training framework so that they are available to all collaborators. Hadley is also involved in calibrating the algorithm performance for flavor tagging.