Fabio Nonato
de Paula
Building AI infrastructure at scale
Senior Manager, Applied Science @ AWS
Building AI infrastructure at scale
Senior Manager, Applied Science @ AWS
I'm a technical leader at the intersection of AI/ML infrastructure and cybersecurity. Based in San Francisco, I lead Applied Science teams at AWS, focusing on threat intelligence and security applications of machine learning.
I build practical tools that bridge the gap between cutting-edge AI research and production systems. My work spans from Bayesian calibration methods and reinforcement learning to modern LLM tooling like MCP servers for local RAG systems.
When I'm not scaling ML infrastructure, I enjoy building developer tools, contributing to open source, and occasionally making games.
Bayesian calibration using TensorFlow Probability
Local FAISS vector store as an MCP server - drop-in local RAG for Claude / Copilot / Agents
Local GraphRAG MCP server for knowledge graph-based retrieval
Cross-platform driver for USB macro keyboard with rotary encoder
Live transcription services from the comfort of your Python interpreter
SageMaker multi-model inference implementation for efficient deployment
Building and scaling ML training pipelines, distributed systems, and inference optimization at AWS scale.
MCP servers, vector stores, local AI tooling. Practical solutions for retrieval-augmented generation.
Threat intelligence, security ML applications, and infrastructure protection at enterprise scale.
Bayesian methods, model calibration, TensorFlow Probability. Uncertainty quantification in production.