Jose Hugo Garcia Aguilar

Expert in modeling multidimensional systems using HPC

Computational physicist with broad experience in HPC algorithms in multi-core and GPU-CUDA platforms, data processing, machine learning, and physics. His scientific background provides him with tools for carrying nonroutine tasks requiring social intelligence, complex critical thinking, and creative problem-solving, skills forged over years of scientific research on numerical methods and improved by working as part of different multidisciplinary projects.
 

Postdoctoral Researcher.  Catalan Institute of Nanoscience and Nanotechnology, Spain  ( 2016 – Current).

• Development and maintenance of a Linux  C/C++ application for sparse and dense linear algebra matrix-vector.
• Development of an algorithm for efficient multi-thread, multi-core and GPU-CUDA parallelization.
• Development of multiple Python post-processing tools  using NumPy, Keras, and Matplotlib libraries.

Postdoctoral Researcher.  International Center For Theoretical Physics – SAIFR, Brazil. ( 2015 – 2016 ).

• Development of an application in C/C++ for multiparametric fitting of non derivable functions and sparse linear algebra using CUDA-C and Python.

Cluster Manager and Software Developer. 
Federal University Of Rio de Janeiro, Brazil (2011 – 2014).

• Maintenance of a heterogeneous computing cluster consisting in hundred of nodes.
• Design of a new algorithm for computing quantum properties of materials using GPUs and CUDA-C.

Cluster Manager. Central University of Venezuela, Venezuela  (2007 – 2011).

• Building and maintaining a 10 nodes clusters using Rocks-OS.

• PhD in Physics from the Federal University of Rio de Janeiro (2015). Thesis: "The kernel polynomial method for quantum transport in disordered systems".
• Degree in Physics from the Central University of Venezuela (2010). Thesis: "Sincronización y control en automatas celulares".