- 📊 Data Loading Dashboard: Real-time corpus overview with document statistics, health metrics, and vector database status
- ❓ Question Groups Overview: Side-by-side comparison of client-provided/approved queries (with support for LLM-generated vs RAGAS-generated questions when necessary) with role breakdown
- ⚙️ Experiment Configuration: Interactive experiment setup with real-time WebSocket streaming and progress tracking
- 📈 Analysis Results Dashboard: Enhanced 3-level analysis with collapsible sections, quick actions, role-based insights, and comprehensive filtering
- 🔍 Gap Analysis Dashboard: NEW! Domain-agnostic content gap detection with practical improvement strategies, role-specific recommendations, and implementation tracking
- 🗺️ Interactive Data Visualization: Multi-perspective hexagonal heatmaps with coverage analytics, unretrieved chunk detection, and smart performance insights
- 🔍 Real Document Processing: Loads CSV and PDF files using LangChain with intelligent chunking (750 chars, 100 overlap)
- 🗃️ Persistent Vector Storage: Qdrant database with cosine similarity search, automatic collection management, and health monitoring
- 📏 Quality Score System: Normalized 0-10 scale with consistent thresholds (GOOD ≥7.0, DEVELOPING ≥5.0, POOR <5.0) and color-coded indicators
- 🆔 Chunk Traceability: Qdrant UUID capture for enhanced debugging and search result analysis with clickable chunk IDs
- 🗺️ Interactive Visualization: D3.js-powered hexagonal heatmaps with multiple perspectives (Chunks-to-Questions, Chunks-to-Roles), advanced analytics, and improved layout.
- 📊 Coverage Analytics: Comprehensive chunk utilization tracking with Unretrieved chunk detection and impactful stats on coverage and performance.
- 👥 Role-Based Analysis: Complete role integration across visualization and analysis workflows
- ⚡ Performance Optimization: Caching, memoization, and optimized rendering for smooth user experience
- 📡 Real-time Communication: WebSocket streaming for live experiment progress updates
- 🧠 Domain-Agnostic Gap Analysis: Rule-based content gap detection that works with any dataset without modification
- 💡 Practical Improvement Strategies: Role-specific recommendations with internal knowledge mining, external research, and LLM-generated content approaches
- 🔤 Rich Text Formatting: Markdown bold formatting support in improvement strategies with HTML rendering
- 🏷️ Feature Flags: Rule-based analysis badges and toggleable UI components
- 📋 Comprehensive Logging: User-friendly logging system with development/production modes
- 🎨 Enhanced UX: Collapsible sections, quick actions, smart insights, cross-screen navigation, and context-aware statistics
- 🎯 Responsive Design: Mobile-friendly interface with CSS Grid and Flexbox layouts