Governed AI Assisted FP&A Architecture Using Cloud Data Warehousing and Semantic SQL Views

This work presents a governed AI-assisted Financial Planning and Analysis (FP&A) architecture integrating cloud-native PostgreSQL warehousing, semantic SQL analytics views, and large language model-based natural language querying using AWS Bedrock. The framework explores how enterprise financial reporting workflows can become more scalable, accessible, and conversational through governed semantic layers and AI-assisted analytics pipelines. The architecture emphasizes reusable business logic, scalable reporting workflows, semantic governance, and AI-assisted enterprise decision support. The proposed system architecture integrates synthetic financial data generation, cloud-native warehousing using Neon PostgreSQL, semantic SQL analytics views, AWS Bedrock AI services, and a natural language interface for conversational enterprise financial analysis. Associated implementation resources and architecture artifacts are included alongside this preprint.