Ugo Manzo
Computer Science student · starting a Master's in Distributed Systems in Milan
I'm a Computer Science student who likes to look under the hood: the complexity of operating systems, memory management in C/C++, cloud infrastructure architecture. I find more satisfaction in optimizing resources and making servers talk to each other efficiently than in designing an interface.
I'm finishing my Bachelor's in Salerno and already have my bags packed for Milan, where I'll soon start a Master's in Distributed Systems. To me, programming isn't just a job. It's a continuous logical puzzle to solve with patience and method. That's also why I'm a member of the ACM: I believe in an approach to computer science that pairs curiosity for research with writing clean, well-documented code.
I'm a methodical type: I want to understand why a system behaves the way it does before I start touching it. That's the same approach whether I'm debugging a service in production or breaking down a logistics problem into manageable pieces for a university project.
What I'm working on right now
I'm preparing for the CKAD (Certified Kubernetes Application Developer) certification, and in parallel deepening my knowledge of cloud-native architecture on AWS: networking, access security, and scaling strategies. The goal is to start my Master's with foundations that are solid in practice, not just in theory, built and stress-tested on my own.
Academic background
MSc in Distributed Systems & Cloud
Milan
2027, starting soon
Chosen to specialize in large-scale distributed architectures, data consistency, and cloud-native infrastructure.
BSc in Computer Science
University of Salerno
2023 – present
Focused on operating systems, networking, databases, and software engineering, with my personal interest gradually shifting toward cloud and distributed systems.
Skills
Primary
Systems & low-level
C/C++ and memory management are still where I really understand what's happening underneath an abstraction. I'm comfortable on the command line on Linux (Fedora/RHEL), with Bash and Git as daily tools.
Cloud & infrastructure
Docker and Kubernetes to containerize and orchestrate applications, AWS to design cloud-native architectures, Ansible and Terraform to automate provisioning and configuration instead of repeating it by hand.
Backend & data
Java and Spring Boot for structured, testable backends, Python when I need to iterate quickly. PostgreSQL is my default relational choice, with MongoDB and Redis for cases that need schema flexibility or in-memory speed.
How I work
I isolate the cause before I touch anything: methodical debugging over trial and error. I document technical decisions because I'll be the first one re-reading them, and I'm comfortable in complex systems with lots of moving, interconnected parts.
Secondary & learning
Rust
I'm studying it to explore systems programming with memory-safety guarantees checked at compile time: a natural next step after C/C++, not yet applied in a complete project.
Development environment
I'm equally comfortable on Linux and Windows, and move easily between VS Code, the JetBrains suite (IntelliJ, CLion, PyCharm), and Vim/Neovim depending on the context.
Certifications & affiliations
CKAD: Certified Kubernetes Application Developer
Linux Foundation
In progress
AWS Cloud Architecture Track
Amazon Web Services
In progress
ACM: Association for Computing Machinery
Member
Access to advanced research and a steady reference point for software engineering best practices.
Projects
HeavyRoute
A 3-tier modular monolith built to digitize logistics workflows still run on paper or scattered spreadsheets. I built the logic and data layers in Java/Spring Boot, keeping presentation, business logic, and persistence cleanly separated so the system stays maintainable as it grows.
- 3-tier architecture with a clean split between presentation, logic, and data
- JWT authentication with role-based access control (RBAC)
- Relational modeling in PostgreSQL for non-trivial logistics workflows
E.R.M.E.S.
A Computer Vision research system for decoding human emotions from images. I trained CNN models and integrated Explainable AI (Grad-CAM), because to me a prediction isn't enough if I can't explain why the model decided what it did. Raw accuracy alone isn't good enough in a domain this sensitive.
- CNN models for emotion classification from images
- Explainable AI with Grad-CAM to surface the regions driving each decision
- Prioritized anatomical consistency and transparency over raw accuracy alone
Outside of work
When I switch the monitors off, my brain has a hard time fully stopping. To disconnect without shutting down the logical engine, I turn to analytical, tactile puzzles: solving a Rubik's Cube on muscle memory, a sudoku, or getting lost in crossword clues. I find it relaxing to take a chaotic situation and bring it back to order through precise rules, a bit like debugging just away from the keyboard.
The same pull toward complexity shows up in what I play. I don't enjoy overly guided experiences; I'm drawn to emergent sandboxes, where surviving means mastering ruthless, interconnected systems. Working out the impact of every resource in a world like Cataclysm: Dark Days Ahead, or keeping a reactor's power grid alive amid the chaos of Space Station 13, takes roughly the same clarity and adaptability needed to keep a failing server infrastructure standing.
Contact
I'm about to move to Milan for my Master's, and in the meantime I'm open to internships, projects, or junior roles in cloud engineering, backend, and distributed systems. If you think my profile could be a fit, reach out. I always reply.