ML research engineer focused on AI safety and interpretability. I am currently working at LASR Labs mentored by Stefan Heimersheim (Google DeepMind).
Previously built an AI assistant with long-term memory and tool integrations for an Anthropic researcher, and served as lead software engineer at an Italian startup.
- LASR Labs Research Fellow (Spring 2026)
- ARENA 7.0 alumnus (1st prize at interpretability hackathon)
- 3+ years as a software engineer (lead front-end at an Italian startup)
- MSc Computer Science & Engineering, Politecnico di Milano
- ML4Good AI Safety Bootcamp (EPFL)
- ML/Research: PyTorch, Hugging Face, TransformerLens, model training/fine-tuning (SFT, DPO), linear probes, RL/RLHF
- AI Integration: Claude API, OpenAI API, RAG, agent development, llama.cpp
- Languages: Python, TypeScript/JavaScript, Java, C++
- Web: FastAPI, React, Django, Angular, Spring Boot, Docker
- Email: fornasiere.raffaello@gmail.com
- LinkedIn: raffaello-fornasiere
- Website: rf-98.com


