Giacomo Baldan
  • Technical University of Munich

    Visiting PhD student in Machine Learning

    January 2025 — Present

    • TUM Physics-based Simulation group (Supervisor: Prof. N. Thuerey).
    • Developed a framework to embed physical knowledge in the diffusion process to build stochastic surrogate models leveraging flow matching.
    • TensorFlow, PyTorch, JAX.
  • Argonne National Laboratory

    Argonne Training Program on Extreme-Scale Computing (ATPESC)

    July 2024

    • Computer architectures, mathematical models and numerical algorithms.
    • HPC systems, and methodologies and tools relevant for Big Data applications.
    • C++ and Python.
    • MPI, OpenMP, OpenACC, CUDA
  • Politecnico di Milano

    PhD student in Aerospace Engineering

    November 2022 — Present

    • Rotor Blade Simulation: from High-Fidelity Simulations to Deep Learning Surrogate Models (Supervisor: Prof. A. Guardone).
    • High-fidelity simulations: RANS, hybrid RANS/LES, and LES.
    • ROMs: semi-empirical and machine-learning based.
  • KTH Royal Institute of Technology

    Researcher

    June 2022 — October 2022

    • High-performance computing.
    • Streaming data architectures, ADIOS2, data compression.
    • C++ and Python.
    • MPI, OpenMP, OpenACC, CUDA
  • German Aerospace Center (DLR) - Institute of Aerodynamics and Flow Technology

    Researcher

    September 2021 — May 2022

    • Employed in the framework of NextSim EU H2020 project.
    • Developer of modern algorithms and their highly parallel implementation in the next generation flow solver CODA (CFD for ONERA, DLR and Airbus) for efficient numerical flow simulation on massively parallel computing clusters.
    • C++ and Python.
    • Hybrid MPI/OpenMP parallelization.
  • Università degli Studi di Padova

    Graduation to professional industrial engineer - Italian legislation

    October 2021

    • Graduated with 60/60.
  • Politecnico di Milano

    MSc in Aeronautical Engineering - Aerodynamics Curriculum

    September 2019 — July 2021

    • Graduated with 110/110 cum Laude.
    • Thesis: An innovative scalable Lagrangian particle tracking approach for distributed-memory computation of dispersed phase flows (Supervisor: Prof. A. Guardone).
  • CINECA

    Virtual School on Numerical Methods for Parallel CFD - Two Weeks

    December 2020

    • State-of-the-art methodologies, numerical methods, and codes for hybrid multi-many core HPC clusters.
  • Università degli Studi di Padova

    BSc in Aerospace Engineering

    October 2016 — July 2019

    • Graduated with 110/110 cum Laude.
    • Thesis: Adaptive mesh refinement algorithms in diffusion problems (Supervisor: Prof. F. Picano).