Data Pipeline Orchestration with Dagster

Context

The growth of analytical infrastructure and the need to integrate data from multiple complex transactional sources (such as ERPs on SAP HANA and storage and logistics systems on SQL Server / Intralog) required a centralized, auditable, and resilient system.

To solve this, the old model of cron tasks and isolated scripts was migrated to a data architecture fully orchestrated with Dagster at Vinos América, structured through independent gRPC microservices coordinated by Docker containers.

AI Integrations with Model Context Protocol (MCP)

Context

To boost decision-making and empower advanced AI assistants (like Claude or ChatGPT) with the ability to query and analyze information in real time, a microservices ecosystem based on the Model Context Protocol (MCP) was designed and implemented.

This ecosystem acts as a secure, governed, and high-performance bridge between Large Language Models (LLMs) and the corporate Data Warehouse (DWH) at Vinos América.

[!NOTE]
Confidentiality and Intellectual Property Note:
Being a private and strategic enterprise development, the source code, credentials, and the Data Warehouse (DWH) codebase/schemas are not publicly available (not open-source) to protect information security and organization operational rules.