#cv
Saleh Jamali Golzar · HPC · inference acceleration · FPGAs · embedded.
education
Nov 2023 — present
PhD Student, Computer Science
University of Salerno, Fisciano, Italy · supervisor: Prof. Biagio Cosenza
topic: High-Performance Computing — optimization on Intel HW & SW platforms
Sep 2016 — Sep 2019
MSc, Electrical Engineering — Integrated Circuit Design
University of Tabriz, Iran · GPA 17.24 / 20
supervisor: Prof. Ghader Karimian · adviser: Prof. Maryam Shoaran
dissertation: point cloud classifier acceleration on GPUs and FPGAs — GPU/FPGA acceleration of the Dynamic Graph CNN model for learning on point clouds
Sep 2012 — Sep 2016
BSc, Electrical Engineering
University of Tabriz, Iran · GPA 16.40 / 20
supervisor: Prof. Hadi Veladi
thesis: data logger over Wi-Fi (ESP8266, Android app)
publications
journals & conferences
2026
Optimizing the LiGen drug discovery pipeline for Intel Max GPUs
S. Jamali Golzar (first author), L. Carpentieri, A. De Caro, B. Cosenza, D. Gadioli, G. Accordi, G. Palermo, F. Ficarelli, D. Gregori, A. R. Beccari
accepted to PDP'26 — 34th Euromicro Int. Conf. on Parallel, Distributed & Network-Based Processing
2025
Demystifying power-of-two quantization: benchmarking inference on AVX and RVV
S. Jamali Golzar (first author), G. Pagano, B. Cosenza
accepted to ITADATA'25-Workshops · Scientific HPC in the pre-Exascale era (2nd ed.)
2022
DGCNN on FPGA: acceleration of the point cloud classifier using FPGAs
S. Jamali Golzar (first author), M. Shoaran, G. Karimian, M. Fattahi
Circuits, Systems, and Signal Processing (Springer) · doi.org/10.1007/s00034-022-02179-0
experience
Apr — Sep 2026
Visiting PhD Student
TU Wien, Vienna, Austria · Parallel Computing Lab · Prof. Sascha Hunold · 6 months
Sep 2025 — Mar 2026
Intern
E4 Computer Engineering SpA, Italy · 6 months
supervisors: Daniele Gregori (E4) · Federico Ficarelli (CINECA)
Oct 2017 — Aug 2018
Embedded & IoT Systems Design
Ava Mechatronics, Tabriz, Iran · part-time
Jun — Aug 2016
Extern
Ava Mechatronics, Tabriz, Iran
skills
hardware
fpga
VerilogHLS C++OpenCLAWS F1
fpga targets
Virtex UltraScale+ VU9PZynq7020Spartan-3Xilinx CPLDs
sch / pcb
Altium DesignerKiCad
mcus
NXP LPC176xST STM8ST STM32Atmel ATmegaAtmel AT91Sam7Cypress FX2
wireless
Espressif ESP8266Espressif ESP32Nordic nRF24TI CC3200
sensors
IMU MPU6050high-G Freescale MMA65XX
rtos / hal
FreeRTOSKeil RTXCMSISESP-IDF
toolchains
open sourceKEILIAR
software
gpgpu
CUDAOpenCLSYCLoneAPI
isas
AVX2AVX-512RISC-V RVV-1.0
ml / dl
inference accelerationONNXPyTorchTensorFlowIREETVM
model compression
graph rewritingquantizationpruning
computer vision
classificationsemantic segmentationobject detection
symbolic
SymPySymEngine
misc
tooling
gitbashLinuxArchlinuxUbuntuAlmaLinux
languages
JavaPythonC++C#.net
selected courses
2016
VHDL · VLSI
University of Tabriz · 18/20 · 19/20
2015
Microprocessors · Computer Architecture
University of Tabriz · 20/20 · 19.5/20
2012
Logic Circuits
University of Tabriz · 20/20
2016
Machine Learning
Stanford University · Coursera (online)
certificates
2025
Multi-GPU Programming Bootcamp
EuroCC2 · online
languages
native
Azeri · Persian
english
TOEFL iBT 102 · UNISA C1