What problem does it solve? Building a cohesive financial-services Databricks proof of concept requires coordinating governed data, predictive ML, GenAI, and a business presentation into one evidence-backed story, which is difficult to sequence and validate without a shared contract. ## Core Features & Use Cases - Four-Layer Orchestration: Sequentially builds a governed data foundation, predictive ML scoring, GenAI activation, and a business presentation layer connected by one trace identifier. - Evidence-Based Manifest: Instantiates a build-demo-pitch manifest where every layer publishes artifacts, evidence, assumptions, and risks with controlled statuses. - Story and Pitch Generation: Produces a five-slide outline, timed demo script, discovery questions, objection responses, and a 30-day pilot plan. - Independent Review: Applies a 100-point readiness rubric with automatic blockers before rehearsal and packaging. - Use Case: A solutions architect preparing a fraud-investigation demo carries one transaction_id from governed Delta tables through an MLflow-scored model, a cited GenAI recommendation, and a business dashboard, then delivers a rehearsed pitch with validated fallback paths. ## Quick Start Ask the agent to build an FSI fraud-investigation demo and pitch from this repository template targeting Azure Databricks with a two-week timebox.
