Hi there, I’m @sirius-22 ✨🙃
Distributed Actor-Based System for Autonomous Logistics.
- Tech: java, Prolog, Gradle, Docker, Raspberry Pi.
- Focus: Designed an asynchronous system using the Actor Model to coordinate a "cargo warehousing and storage services" with autonomous robot (DDR).
- Highlights: Implemented Clean Architecture, comprehensive unit testing, and deployed on embedded hardware.
An intelligent agent for the strategic board game Tablut.
- Tech: Java, AI Search Algorithms, DNN, pytorch.
- Focus: Implemented Minimax with Alpha-Beta Pruning and custom heuristics to compete in the University of Bologna AI challenge.
- Tech: RISC-V, PyTorch, BLIS, OpenBLAS, Python, Linux.
- Focus: Performed a quantitative performance analysis of GPT-2 and DeepSeek models on a many-core RISC-V architecture. Optimized PyTorch inference by configuring and comparing different BLAS backends (OpenBLAS vs. BLIS) to enable CPU vectorization. Analyzed scaling performance across 64 cores, measuring metrics like token generation rate and inference latency to identify architectural bottlenecks.
🎓 B.Sc. Thesis: Researched Machine Learning strategies for protecting critical infrastructure (ICS) from cyber-threats.
- 🔭 Tech Focus: I love exploring AI Trends, Computer Vision, and Software Architecture.
- 🎲 Gaming: Huge fan of Dungeons & Dragons and Tabletop Games (ask me about my latest campaign!).
- 📚 Culture: I enjoy Anime & Manga and staying curious about new tech.
- 🧩 Challenge: I actively practice Coding Challenges to keep my problem-solving skills sharp.
