FastData Platform Specification

Version: 3.0 Last Updated: 2026-05-10

1. Overview

FastData is an AI-native data analysis platform with a ChatGPT-style conversational interface. It enables users to upload CSV/Excel files and receive instant AI-powered analysis, visualizations, and insights through natural language interactions.

1.1 Key Features

Conversational Interface - Chat with your data using natural language
Real-time Analysis - Automatic data profiling, correlation analysis, and insight generation
Domain Detection - Automatic detection of financial, crypto, ecommerce, medical, and marketing domains
Domain-Specific Analytics - Specialized algorithms for each detected domain
Offline Mode - Full client-side analysis when server is unavailable
Export Capabilities - Multiple export formats including PPTX, JSON, and ZIP packages

2. Architecture

2.1 Frontend (React + TypeScript)

src/
├── components/
│   ├── Chat/              # Main chat interface
│   ├── Sidebar/          # Conversation history
│   ├── FileUpload/       # Drag & drop file handling
│   ├── Dashboard/        # Interactive dashboard
│   ├── InsightsWidget/   # Insights display
│   ├── CorrelationsWidget/ # Correlation matrix
│   ├── ChartsWidget/     # Chart visualizations
│   ├── Copilot/          # AI chat assistant
│   └── CopilotPanel/     # Expandable AI panel
├── engine/
│   ├── parser.ts         # CSV/Excel parsing
│   ├── profiler.ts       # Data profiling
│   ├── correlations.ts   # Correlation engine
│   ├── insightGenerator.ts # Insight generation
│   ├── chartRecommender.ts # Chart recommendations
│   ├── reportGenerator.ts # Report generation
│   ├── exportEngine.ts   # Export functionality
│   ├── slideGenerator.ts # Slide generation
│   ├── copilot.ts        # Client-side copilot
│   ├── aiEngine.ts       # AI integration
│   └── domains/          # Domain-specific analytics
│       ├── financial.ts   # Financial indicators
│       ├── crypto.ts     # Crypto analysis
│       ├── ecommerce.ts  # Ecommerce analytics
│       ├── medical.ts    # Medical/clinical analytics
│       └── marketing.ts  # Marketing analytics
├── services/
│   └── api.ts            # Backend API client
├── stores/
│   └── appStore.ts       # Zustand state management
└── types/
    └── index.ts          # TypeScript definitions

2.2 Backend (Express + TypeScript)

server/
├── src/
│   ├── index.ts          # Express server entry
│   ├── config/           # Configuration
│   ├── routes/
│   │   ├── upload.ts     # File upload & analysis
│   │   ├── copilot.ts    # AI copilot endpoints
│   │   └── reports.ts    # Report generation
│   ├── services/
│   │   ├── fileProcessor.ts    # File parsing
│   │   ├── analyzer.ts        # Main analyzer
│   │   ├── domainDetector.ts  # Domain detection
│   │   ├── insightEngine.ts   # Server-side insights
│   │   ├── copilotService.ts  # Copilot logic
│   │   ├── aiOrchestrator.ts  # AI orchestration
│   │   ├── narrativeEngine.ts # Narrative generation
│   │   ├── semanticTagger.ts  # Semantic tagging
│   │   ├── exportService.ts  # Export service
│   │   └── reportGenerator.ts # Report generation
│   └── types/
        └── index.ts      # Server types
└── package.json

3. API Endpoints

3.1 Upload Routes (`/api/upload`)

Method	Endpoint	Description
POST	`/api/upload`	Upload CSV/Excel file
POST	`/api/upload/:id/analyze`	Run analysis on dataset
GET	`/api/upload/:id`	Get dataset info
GET	`/api/upload/:id/insights`	Get filtered insights
GET	`/api/upload/:id/anomalies`	Get detected anomalies
GET	`/api/upload/:id/correlations`	Get correlations
GET	`/api/upload/:id/charts`	Get chart configs
GET	`/api/upload/:id/report`	Generate report
POST	`/api/upload/:id/export`	Export analysis
DELETE	`/api/upload/:id`	Delete dataset

3.2 Copilot Routes (`/api/copilot`)

Method	Endpoint	Description
POST	`/api/copilot/query`	AI copilot query
POST	`/api/copilot/explain/:insightId`	Explain insight
POST	`/api/copilot/narrate`	Generate narrative
GET	`/api/copilot/intents`	List available intents

3.3 Reports Routes (`/api/reports`)

Method	Endpoint	Description
GET	`/api/reports/styles`	List report styles
POST	`/api/reports/generate`	Generate report
POST	`/api/reports/generate-pptx`	Generate PowerPoint
POST	`/api/reports/generate-pdf`	Generate PDF
GET	`/api/reports/template`	Get template

4. Domain-Specific Analytics

4.1 Financial Domain

Feature	Description
RSI	Relative Strength Index calculation
MACD	Moving Average Convergence Divergence
Bollinger Bands	Price volatility bands
Drawdowns	Portfolio drawdown analysis
Financial Metrics	Returns, volatility, Sharpe ratio, Sortino ratio
Pattern Detection	Trend pattern recognition

4.2 Crypto Domain

Feature	Description
Wallet Analysis	Address clustering, risk scoring
Gas Optimization	Gas price recommendations
Token Metrics	Price, volume, market cap analysis
Staking Analysis	APY, rewards calculation
Portfolio Analytics	Allocation, P&L tracking
Rug Pull Detection	Risk scoring for tokens

4.3 Ecommerce Domain

Feature	Description
RFM Analysis	Recency, Frequency, Monetary segmentation
Cohort Analysis	Customer retention analysis
Basket Analysis	Market basket analysis, association rules
Customer Lifetime Value	CLV prediction
Cross-sell Recommendations	Product recommendations

4.4 Medical Domain

Feature	Description
Survival Analysis	Kaplan-Meier curves, hazard ratios
Clinical Distributions	Normal/skewed/bimodal detection
Clinical Trials	Phase analysis, efficacy metrics
Diagnostic Accuracy	Sensitivity, specificity, PPV, NPV
Patient Segmentation	Cluster-based segmentation
Biomarker Analysis	Correlation with outcomes

4.5 Marketing Domain

Feature	Description
Funnel Analysis	Conversion tracking
Attribution Models	First-touch, last-touch, linear, etc.
Customer Journey	Path analysis
Channel Performance	ROAS, CPA, CTR metrics
Marketing Mix Modeling	MMM with saturation effects
CLV Calculations	Customer lifetime value

5. Advanced Analytics Engine

5.1 Forecasting

Moving average projections
Seasonality detection
Anomaly-aware predictions
Confidence intervals
Ensemble methods support

5.2 Anomaly Detection

Statistical outliers (Z-score, IQR)
Regime shift detection
Contextual anomalies
Pattern-based detection

5.3 Clustering

K-means clustering
Hierarchical clustering
DBSCAN support
Cluster profiling

5.4 PCA (Principal Component Analysis)

Dimensionality reduction
Variance explained
Component interpretation

5.5 Dataset Comparison

Side-by-side metrics
Difference analysis
Trend comparison

5.6 Time Series Decomposition

Trend extraction
Seasonality detection
Residual analysis

6. Premium Visual Components

6.1 Animated Metrics

Number counting animations
Trend indicators
Sparkline charts
Gauge displays

6.2 Chart Features

ECharts integration
Interactive tooltips
AI-generated insights on hover
Domain-specific styling

6.3 Dashboard Layout

Widget-based layout
Drag-and-drop positioning
Responsive grid system
Theme support (light/dark)

7. Export Formats

Format	Description
PPTX	PowerPoint presentations
JSON	Full analysis package
PDF	Markdown reports
HTML	Interactive dashboards
ZIP	Complete analysis bundle

7.1 Export Contents

Charts (PNG/SVG)
Reports (markdown)
Slides (PPTX)
Data summary
Original data

8. Enterprise Features (Planned)

8.1 Multi-tenancy

Organization isolation
Role-based access control
Custom workspaces

8.2 API Rate Limiting

Request throttling
Quota management
Usage tracking

8.3 Usage Analytics

API usage metrics
Feature adoption tracking
Performance monitoring

8.4 Custom Branding

White-label support
Custom themes
Logo customization

9. Tech Stack

Frontend

Framework: React 19 + TypeScript
Build Tool: Vite 8
Styling: Tailwind CSS v4
Charts: Apache ECharts 6
State: Zustand 5
Animation: Framer Motion 12
Icons: Lucide React

Backend

Runtime: Express 4
AI: Anthropic SDK, OpenAI SDK
Queue: Bull (Redis)
Export: PptxGenJS

10. Configuration

Environment Variables

Variable	Description	Default
`PORT`	Server port	3001
`NODE_ENV`	Environment	development
`OPENAI_API_KEY`	OpenAI API key	-
`ANTHROPIC_API_KEY`	Anthropic API key	-
`OPENROUTER_API_KEY`	OpenRouter API key	-
`CORS_ORIGIN`	CORS origin	*
`MAX_FILE_SIZE`	Max upload size	50MB

11. Error Handling

Client-Side Fallback

When the backend is unavailable:

Connection status indicator shows "Offline Mode"
Client-side analysis engine takes over
All core features remain functional
Visual indicators show offline state

API Error Responses

{
  "error": "Error message",
  "code": "ERROR_CODE",
  "details": {}
}

12. Testing Checklist

Server Endpoints

Frontend Integration

API service connected
Copilot component added to Chat
Offline fallback works
Connection status indicator

Domain Analytics

Financial (RSI, MACD, Bollinger, drawdowns)
Crypto (wallet, gas, tokens)
Ecommerce (RFM, cohort, basket)
Medical (survival, clinical, diagnostic)
Marketing (funnels, attribution, journey)

13. Future Enhancements

Advanced Analytics Engine
- Ensemble forecasting models
- Contextual anomaly detection
- Advanced clustering algorithms
- Real-time dataset comparison
Premium Visual Components
- Animated metric displays
- AI-powered chart tooltips
- Dashboard layout generator
- Custom visualization builder
Export Enhancements
- Interactive HTML dashboards
- PDF reports with embedded charts
- Shareable analysis links
- Email report scheduling
Enterprise Features
- Multi-tenant architecture
- API rate limiting
- Usage analytics dashboard
- Custom branding options

14. License

MIT License - Built with ❤️ for the future of data analytics.

FAST DATA - Your AI Data Analyst in a Chat

FilesExpand file tree

SPEC.md

Latest commit

History