Mission: Data Analyst – Find Zava's Hidden Product-Quality Crisis

Mission/open goal Description

My goal was to identify the highest-risk product-quality issue in Zava's PromptathonDb by combining sales performance, support ticket analysis, customer satisfaction, support chat themes, and AI-powered vector similarity search. I used SQL MCP tools and GitHub Copilot Agent to investigate quantitative and qualitative evidence, identify recurring complaint patterns, and recommend the product requiring immediate business intervention.

Harness and model

GitHub Copilot Chat (Agent Mode), using the default model configured by GitHub Copilot.

Turn-by-turn journey

MICROSOFT.docx

Completion

• Yes, the agent completed the mission or goal.
• No, the agent did not complete the mission or goal.

Bonus work

No additional bonus mission was completed. The submission includes enhanced documentation (objective, methodology, and AI workflow) in addition to all required mission tasks.

Start with the MICROSOFT.docx attachment and the mission description, then review the referenced SQL MCP and GitHub Copilot Agent workflow. A completed submission should document the combined sales, support, satisfaction, chat-theme, and vector-similarity analysis and identify the highest-risk product.