Master's student in Language Technologies at University of Turin. Graduating soon.
I work as an AI/ML engineer, designing and building systems that take models from research into production. My background in linguistics shaped my focus on NLP, LLMs, and RAG — the core of my current Master’s research at the University of Turin. A recurring thread in my work is spotting where standard methods break on edge cases — like fixing tokenizer vocabulary gaps for crypto terminology, or introducing a new scoring metric to separate agent from patient roles in slur reclamation — and building the fix myself rather than working around it. I have worked across industry and academic settings in Italy, Germany, and Iran.
The Noise of the Information Age
graph LR
%% 1. The World of Noise
subgraph Context [ ]
A(Data Bombardment)
B(Infinite Inputs)
C(Fragmented Signals)
end
%% 2. The Tinkerer's Process
subgraph Process [Selection & Placement]
D{Deciphering}
E[Selection: Finding Components]
F[Placement: Weaving Connections]
end
%% 3. The Result
subgraph Outcome [The Coherent Whole]
G((CLARITY))
H[Insight]
I[Orchestrated Systems]
end
%% Connections
%% By connecting A, B, and C to D individually,
%% Mermaid naturally stacks them vertically to save space.
A -.-> D
B -.-> D
C -.-> D
D -->|Filtering| E
E -->|Arranging| F
F ==>|Transformation| G
G --> H
G --> I
%% Styling
style G fill:#f9f,stroke:#333,stroke-width:4px
style D fill:#fff,stroke:#333,stroke-width:2px
style E fill:#fff,stroke:#333,stroke-width:2px
style F fill:#fff,stroke:#333,stroke-width:2px
style Context fill:#f9f9f9,stroke:#ddd,stroke-dasharray: 5 5