# Eugenio "Jay" Zuccarelli

> Eugenio "Jay" Zuccarelli is an AI-native operator who builds agentic AI systems and ships machine learning models in production, working across the full lifecycle from research and modeling to deployment.

## Identity

- Full name: Eugenio Zuccarelli
- Also known as: Jay Zuccarelli
- Website: https://jayzuccarelli.com/
- GitHub: https://github.com/jayzuccarelli
- LinkedIn: https://linkedin.com/in/jayzuccarelli
- X (Twitter): https://x.com/jayzuccarelli

## Focus areas

- Applied AI: agentic AI, generative models, production AI, retrieval-augmented generation, evals, RLHF
- ML modeling: supervised, unsupervised, and reinforcement learning; feature engineering; model selection; hyperparameter tuning
- Data science: experimentation, causal inference, forecasting, decision systems

## Stack

Python, PyTorch, TensorFlow, AWS, GCP, SQL, Spark, Docker, Kubernetes, Git, Linux, Bash.

## Open-source projects

- saccade: Open harness for proactive ambient agents that watch, listen, and only speak up when it's worth your attention. Local model runs continuously on-device for privacy; remote model only fires for harder calls.
- [autofill](https://github.com/jayzuccarelli/autofill): AI-powered form autofill agent built on browser-use. Describe yourself once, point it at any web form, and it fills every field automatically; the user reviews and submits manually.
- [ha-memory](https://github.com/jayzuccarelli/ha-memory): Persistent memory for Home Assistant Assist, stored in plain markdown. No vector DB and no embeddings — works with Anthropic Claude, OpenAI, and any LLM conversation agent.