Andrea Antonaci

I’m a Cloud & AI Platform Engineer specializing in production agentic AI platforms on AWS (Bedrock, Strands Agents, MCP, RAG). I help teams move from prototypes to platforms they can operate with clear architecture, infrastructure as code, and costs you can defend to a CFO.

In my current role, I designed and built multi-agent systems on Amazon Bedrock using the Strands Agents SDK, integrating RAG pipelines with source-control and knowledge tools via read-only MCP servers. I architected retrieval and memory layers with Bedrock Knowledge Bases and S3 Vectors, and deployed defense-in-depth AgentCore services (FastAPI) as containerized workloads.

I led architecture and TCO for a self-service agentic AI platform, where new domain agents are defined declaratively and backed by Terraform plus an evaluation harness. I also built serverless AI pipelines (Lambda, Step Functions, Textract, DynamoDB, S3 Vectors) provisioned end-to-end with Terraform and GitHub Actions.

Previously, as a Cloud Architect / Project Manager, I delivered cloud architectures primarily on Azure (with selected AWS and GCP) and took documental RAG solutions from proof-of-concept to production. I drove FinOps cost optimization and governance across Azure environments and project-managed multiple web-application builds from requirements through release.