designing-real-world-ai-agents-workshopMCP Server37/100 via “workflow test scripts and batch processing automation”
Hands-on workshop: Build a multi-agent AI system from scratch — Deep Research Agent + Writing Workflow served as MCP servers. Includes code, slides, and video
Unique: Combines Python scripts with Makefile-based task orchestration, enabling both programmatic control (for CI/CD) and simple command-line invocation (for developers). Scripts handle full workflow automation including dataset loading, result collection, and metric aggregation.
vs others: More accessible than custom Python orchestration because Make commands are simple and discoverable, and more flexible than hardcoded test suites because scripts are parameterized for different datasets and profiles.