Summary of GSoC 2025 Proposals

Click here for an overview of the GSoC program at HSF.

Full List of Proposals

ATLAS: Precision Recovery in Lossy-Compressed Floating Point Data for High Energy Physics
baler: The rise of the machine (learning) in data compression for high energy physics and beyond
baler: Probabilistic circuit for lossless HEP data compression
BioDynamo: Agent-Based Simulation of CAR-T Cell Therapy Using BioDynaMo
Patatrack: Development of an auto-tuning tool for the CLUEstering library
CernVM-FS: Evaluate Distribution of ML model files on CVMFS
Clad: Implement and improve an efficient, layered tape with prefetching capabilities
Clad: Enhancing LLM Training with Clad for efficient differentiation
Clad: Enable automatic differentiation of OpenMP programs with Clad
Clad: Enable automatic differentiation of OpenMP programs with Clad
Clad: Integrate Clad to PyTorch and compare the gradient execution times
Clad: Enable automatic differentiation of C++ STL concurrency primitives in Clad
Clad: Support usage of Thrust API in Clad
Constellation: Extending the User Interface
CppInterOp: Implement CppInterOp API exposing memory, ownership and thread safety information
Ganga: Incorporate a Large Language Model to assist users
Ganga: Implement a deprecation system to keep code up to date
Geant4: Geant4-FastSim - Data Representation Optimisation for Generative Model-based Fast Calorimeter Shower Simulation
ROOT: Using ROOT in the field of genome sequencing
CICADA: Highly Granular Quantization for CICADA
HSFCondDB: Intelligent Log Analysis for the HSF Conditions Database
CppInterOp: Interactive Differential Debugging - Intelligent Auto-Stepping and Tab-Completion
JuliaHEP: Julia Interfaces to HepMC3
MCnet: MCnet/MCviz - graph and 3D-viewer tools for simulated particle collisions
MCnet: MCnet/OpenData - tools and exercises for open-data exploration with MC simulations
ML4EP: Integrating Support for Google XLS in HLS4ML
ML4EP: Optimizing Model Splitting in hls4ml for Efficient Multi-Graph Inference
ROOT: RNTuple in JSROOT
Rucio: Rucio WebUI Revamp
BEAD: Background Enrichment augmented Anomaly Detection (BEAD) for new physics searches at LHC
SMARTHEP: Estimating the energy cost of ML scientific software
ML4EP: TMVA SOFIE - GPU Support for Machine Learning Inference
ML4EP: TMVA SOFIE - HLS4ML Integration for Machine Learning Inference
ML4EP: TMVA SOFIE - Enhancing Keras Parser and JAX/FLAX Integration
Xeus-Cpp: Implementing Debugging Support
Xeus-Cpp: Enable GPU support and Python Interoperability via a Plugin System