Nobile Engineering

About Nobile Engineering

Nobile-engineering comes from the effort of Andrea Nobile to bring expertise in the broad field of computer vision algorithms and embedded applications to demanding customers.

Andrea Nobile has a Ph.D. in numerical physics with a long experience as researcher. In particular he is an expert of computer vision, numerical algorithms and implementation of high performance, low latency code in C/C++, ASM and intrinsics (x86, ARM, Power).

Andrea is passionate about computer vision, machine learning, data science, embedded development and software development in general.

Andrea can work on heterogeneous platforms extracting the maximum performance by working in a scientific manner that is uncommon among practitioners, he is comfortable with accelerators, having experience in HW accelerator design as HW architect.

Due to his scientific research background in numerical physics and HPC hardware and software, he is the right person to design and implement cutting edge, absolute high-performance algorithms, in particular demanding computer vision and machine learning algorithms. He has working knowledge and experience using Python and C++ with Caffe, PyTorch, MXNET, Tensorflow, XGBOOST, LibLINEAR, LibSVM, and the most powerful machine learning tools. He has experience of the internals of Caffe. As hobby he is writing a CNN runtime that uses a variety of tricks to speed-up execution of CNNs.

He is an absolute HPC expert. He had critical roles in the QPACE project including: main algorithms design and implementation, scientific code porting, design and implementation of the Linux kernel driver for the 3D torus network, design and implementation of the user-space 3D torus network library, testing and debugging of the 3D network HW.

He furthermore worked in cooperation with IBM on cutting-edge projects including evaluation of the Blue-Gene/Q prototype and the in-memory-processing research architecture AMC.

He is an expert about code parallelization using SIMD, multithreading (pthread, OPENMP) and network using MPI and other APIs. He is confident with low-level programming, directly interacting with the HW.

Finally he is also the author and administrator of this website. It runs on an cloud Ubuntu VM