I write embedded software: mostly C++ on ESP32s, mostly for things that are supposed to fly.
I was the project manager of the avionics software for the first two rockets of Aurora Rocketry, the University of Bologna's student rocketry team, and I do research on moving large amounts of data over LoRa without wasting airtime. I'm doing an MSc in Computer Science and Engineering in Bologna, and I spent the spring of 2026 on Erasmus in Reykjavík.
During the day I'm a software developer at a company that makes enterprise software. Not as thrilling as rockets, I know. The long-term plan is to have some code of mine running in space before I turn 32.
Aurora RocketryFlight software and telemetry for Borealis and Nemesis
Aurora Rocketry is the University of Bologna's student rocketry team. I was the project manager of the avionics software for its first two rockets, Borealis in 2024 and Nemesis in 2025, with a team of seven. The flight computer of Borealis and its LoRa telemetry to the ground became my bachelor's thesis. The round white board below is the Nemesis one.
The Nemesis avionics board, with the flight state machine on screen. Photo: Aurora Rocketry.
A LoRa frame holds about 255 bytes, so a flight log has to be sent in pieces. The usual approach waits for an ACK after every piece. LAMP sends a burst and gets back a single bitmap of what's missing. In ns-3, in an urban channel at SF7, that gives 4.6 km of range at 90% delivery, against 3.4 km for stop-and-wait. I'm writing the paper with Prof. Andrea Piroddi.
header 9 Bpayloadcrc16 2 B
stop-and-wait
0ack1ack2ack3ack4ack5ack
LAMP
01234567sack 111011113
Top: one chunk on air. Below: chunk 3 is lost. Stop-and-wait pays an ACK per chunk; LAMP pays one bitmap per burst and resends only chunk 3.
I reverse-engineered the RF modules of smart cinema seats and wrote a simulator of the whole system, so the IoT platform could be tested without the seats. Integration testing got 60% faster.
.NET 8 · Python · RF · code is private
River Monitoring SystemEmbedded Systems & IoT course
An ESP32-S3 with a sonar measures the water level and publishes it over MQTT. A Java service decides how far to open the channel valve, and an Arduino moves it with a servo and shows the state on an LCD. There's a Python dashboard on top.
Circles, in parallelHigh Performance Computing course
The same circle-intersection simulation in C, parallelised twice: with OpenMP on shared memory and with MPI across processes, with strong and weak scaling measurements for both.
Federated Lightweight ChatReplication, on purpose the hard way
A group chat I wrote to study replication. Every user gets a container with a full SQLite copy of every room. Nodes converge by gossip and agree on message order with vector clocks, and a node keeps accepting messages when it's cut off or Redis is down.
A Spark pipeline that joins NOAA weather data with ENTSO-E grid data, to look at things like how temperature drives demand or how much cloud cover costs solar.
Each skill loads the design constraints of one well-known engineer, based on what they've published: Carmack, Lamport, Muratori, Ousterhout and others. The part I like is the review mode, where two opposing ones critique the same code.
Borealis and Nemesis: flight software and telemetry, team of 7. Details
2024
Embedded software intern, SPOT. Digital Made Simple
RF reverse engineering and the OKIN simulator.
2023 – 2024
Software Developer, UniBo Motorsport
Warehouse software for the Formula Student team: database pooling and QR scanning.
Education
2025 – now
MSc, Computer Science and Engineering
University of Bologna
2026
Erasmus+ semester
Reykjavík University
2021 – 2025
BSc, Computer Science and Engineering
University of Bologna, thesis on rocket telemetry over LoRa
Tools
Mostly C and C++ on ESP32 (ESP-IDF, FreeRTOS, PlatformIO, Unity for tests), and Python for everything around them. I also use Java, C#, Go, SQL, Docker and Linux, and I've written OpenMP, MPI and CUDA code for HPC courses.
On GitHub
1069 contributions in the last 12 months, on 88 different days.
Writing
I write on my blog, in Italian and English, about work, projects and things I'm learning.