Electronic Engineering MSc

Gianmarco
Fortunelli

MSc student in Electronic Engineering interested in computer architecture, RTL design, ASIC, FPGA-based acceleration, and hardware security. I care about how design choices at the electronic level shape performance, power, and security — from arithmetic units and FPGA pipelines to cache-timing side channels.

Computer Architecture RTL Design ASIC FPGA RISC-V Hardware Security Digital Design
Education
2026 Present

ETH Zurich — Switzerland

Master's Thesis Student

Designing a configurable multi-datatype arithmetic unit for ML workloads, extending the MXDOTP RISC-V ISA extension to support FP16 accumulation and mixed-datatype operations. Supervised by Danilo Cammarata, Gamze Islamoglu, and Prof. Dr. L. Benini.

2025 Present

EURECOM — Sophia Antipolis, France

MSc Internet of Things — Double Degree

Double-degree with Politecnico di Torino. Courses: System Security, Machine Learning and Intelligent Systems, Reinforcement Learning, Software Development. Semester project on cache-timing attacks against T-table AES on RISC-V hardware.

2024 Present

Politecnico di Torino — Italy

MSc Electronic Engineering — Embedded Systems

GPA 29.1/30. Courses: Computer Architectures, Microelectronic Systems, Integrated Systems Architecture, Synthesis and Optimization of Digital Systems, Operating Systems.

2021 2024

Politecnico di Torino / Tongji University

BSc Electronic Engineering — Double Degree

Final grade: 109/110. Double-degree with Tongji University, Shanghai. Final project: FPGA-based digital control of an agro-tech sensor using complex ASM.

Experience
2025 Present

EURECOM — France

Graduate Researcher

Research under Prof. R. Pacalet on hardware security and microarchitectural side-channel attacks on RISC-V. Developed and evaluated a cross-process L1 Prime+Probe attack against T-table AES on a BeagleV Fire / PolarFire SoC — cache-timing leakage, attacker-victim synchronization, noise characterization, and statistical key inference. Work accepted as a poster at RISC-V Summit Europe 2026.

2023 2025

Team RoboTO — Italy

Electronics Division Leader

Led the electronics division, coordinated a student team, managed deadlines, and developed supercapacitor modules for RMNA 2025 in San Diego (DJI Robomaster). Mentored junior members.

2024 2025

University of Perugia — Italy

Research Collaborator

Under Prof. S. Simonetti on biomechanical support systems. Designed and validated a back brace monitoring system using FSR sensors.

2023 2024

Politecnico di Torino — Italy

Teaching Assistant

TA for Computer Science (Python) course, A.Y. 2023/2024. Supported students during programming labs and algorithmic design activities.

2022 2023

Tongji University — China

Undergraduate Researcher

Supervised by Prof. X. Gong. Research on biological signal processing and EEG-based brain-computer interfaces for emotion recognition. Published at IEEE ICNC-FSKD 2024.

Selected Projects

Multichannel Conv1D Hardware Accelerator

RTL accelerator for 1D convolution under strict resource constraints. Output-centric scheduling, tiling, Croc SoC integration, bare-metal C drivers, full RTL-to-GDSII flow. 149.5× CPU speedup. Verilator, Yosys, OpenROAD.

Accelerator · RTL

Low-Latency FPGA Pipeline for UDP Market-Data Processing

Zynq-based FPGA pipeline for MoldUDP64 / ITCH UDP market-data processing. Flat-FSM RTL parser for Ethernet, IPv4, UDP, MoldUDP64, and ITCH Trade messages, connected to a fixed-point decision engine. Deterministic packet parsing and hardware-accelerated decision logic. Verilog, Vivado, Vitis, AXI-Stream, AXI DMA, AXI-Lite, Zynq-7020.

FPGA · Low-Latency

Cache-Timing AES on RISC-V

Cross-process L1 Prime+Probe attack on T-table AES. BeagleV Fire / PolarFire SoC. Cache-set leakage analysis, attacker-victim sync, noise characterization, partial key recovery via statistical inference.

HW Security

Pipelined DLX Processor

Fully pipelined DLX processor in RTL. Full instruction support, branch prediction, hazard management, and forwarding logic. Performance-oriented datapath/control design. VHDL, QuestaSim.

Processor · VHDL

Multi-Vth Power Optimization

TCL-based low-power optimization flow: selective cell swapping to higher-Vth variants while preserving timing, with automated tuning and result aggregation in Synopsys PrimeTime.

ASIC Flow · TCL

Linux Scheduler Experiments

Automated benchmarking of Linux scheduling policies in a QEMU environment: kernel config, workload execution, tracing, latency analysis. Linux 5.4.281, QEMU, BusyBox, Ftrace, GCC.

OS · Systems

PACMAN Embedded Game

Custom Pac-Man on LandTiger LPC1768. Real-time gameplay, joystick input, A* ghost pathfinding, chase/frightened FSM, scoring, sound effects. C, Keil µVision, ARM Cortex-M3.

Embedded · C
Publications & Awards

Publications

RISC-V Summit Europe 2026 — Accepted Poster

From Leakage to Exploitability: Empirical Study of Cross-Process L1 Prime+Probe on RISC-V

G. Fortunelli, R. Pacalet.

IEEE ICNC-FSKD 2024

Phase-Amplitude Coupling of EEG Applied in Music-Induced Emotional Recognition Tasks

M. Zhao, G. Fortunelli, Z. He, X. Gong, A. Cohn.

doi: 10.1109/ICNC-FSKD64080.2024.10702312 ↗

Awards

  • Politong Scholarship — PoliTo–Tongji double-degree
  • Erasmus Scholarship — PoliTo–EURECOM double-degree
  • Progetto Intraprendenti Honors Program — top 200 students
  • Matematica e Realtà National Math Competition — winner, 2021

Languages

  • Italian — Native
  • English — Advanced (C1)
  • French — Basic
  • Chinese — Basic
Technical Profile

HDL / Languages

  • Verilog / SystemVerilog
  • VHDL
  • C
  • Python
  • TCL
  • Bash

EDA / Synthesis

  • Synopsys DC Shell
  • Synopsys PrimeTime
  • QuestaSim
  • Verilator
  • Yosys / OpenROAD
  • Vivado / Vitis

Systems / FPGA

  • Zynq-7020
  • AXI-Stream / AXI DMA
  • RISC-V ISA
  • QEMU
  • GCC toolchain
  • Linux kernel
  • ARM Cortex-M

Research Areas

  • Computer Architecture
  • RTL Design
  • ASIC Flow
  • Hardware Security
  • Side-Channel Analysis
  • ML Hardware
Contact

Let's connect.

Available for research collaboration and internship opportunities. The fastest route is email or LinkedIn.