About Me

Why I build AI products

I’m a Rome-based MLOps engineer who thrives at the intersection of applied research and production systems. I get teams shipping faster by pairing robust infrastructure with pragmatic workflows—from GPU orchestration and data pipelines to safety tooling for customer-facing agents.

Experience

Stip AI · Lead ML Engineer

Sep 2024 – Present · Rome, Italy

  • Built the AI organisation from the ground up, owning hiring, roadmap definition, and delivery.
  • Designed a multi-agent voice platform on LangGraph + LangChain with tight latency budgets.
  • Scaled RAG pipelines through metadata-rich ingestion, automated evaluation, and CI/CD.

Translated · MLOps Engineer & AI Production Lead

Feb 2022 – Sep 2024 · Rome, Italy

  • Led a cross-functional engineering pod maintaining 100+ machine translation models in production.
  • Migrated AI infrastructure to a geo-distributed Kubernetes + SLURM stack, reducing idle GPU time by 30%.
  • Developed an internal annotation platform producing 100k+ high-quality datapoints for proprietary LLMs.

Stip · AI Specialist

Sep 2021 – Feb 2022 · Rome, Italy

  • Shipped customer-care NLP solutions end-to-end—from data preparation to monitoring in production.
  • Implemented reproducible ML workflows with MLflow, DVC, and automated deployment tooling.

University of Rome “Tor Vergata” · NLP Research Scholar

Sep 2020 – Feb 2021 · Rome, Italy

  • Explored rule injection for neural dialog systems and authored an LREC 2022 publication.
  • Collaborated with AlmaWave to evaluate explainability techniques for healthcare diagnostics.

Education

M.Sc. Computer Science · 110/110 cum laude

University of Rome “Tor Vergata” · 2019 – 2021

Thesis: Evaluation of Explainability of Machine Learning Models: ERO-MD Explainable Model for Migraine Diagnosis

B.Sc. Computer Science

University of Rome “Tor Vergata” · 2016 – 2019

Thesis: Automation in semantic annotations and ontology population for United Nations resolutions

Skills snapshot

Python PyTorch LangChain LangGraph TensorRT vLLM Kubernetes Docker SLURM MLflow Airflow GitHub Actions Postgres MongoDB Redis

Recognition & community

EVALITA 2020 · Best System Award

1st place in the SardiStance task at the national Italian NLP competition.

Pi Campus HackParty W21 · 2nd Place

Runner-up in the AI Challenge, delivering a production-ready prototype in under 48 hours.

Student Representative & Hackathon Organizer

Advocated for Computer Science students and curated innovation events at Tor Vergata (2017–2021).

Publications

Combining Transformer-based Architectures for Stance Detection

EVALITA · 2020

Read the paper ↗

Certifications

  • DeepLearning.AI · Introduction to Machine Learning in Production (2021)
  • Udemy · Docker and Kubernetes: The Complete Guide (2021)
  • HackerRank · REST API (Intermediate) (2020)

Languages

  • Italian · Native
  • English · B2