✨ AI Features Overview

Sentio provides powerful AI-driven capabilities that make blockchain data analysis accessible through natural language queries. The AI system can generate SQL queries, create insights and visualizations, and even generate custom processor code to track specific blockchain events.

Key AI Features

🤖 Natural Language Data Querying

Transform plain English questions into complex blockchain data queries:

SQL Generation: Convert natural language into optimized SQL queries
Smart Table Selection: Automatically identify relevant data tables
Query Explanation: Understand what each generated query does
Error Correction: Automatically fix and improve queries

📊 Intelligent Insights & Analytics

Get instant analysis and visualizations of your blockchain data:

Metrics Analysis: Generate time-series queries for KPIs and metrics
Chart Recommendations: Suggest optimal visualization types
Multi-dimensional Analysis: Analyze data across different dimensions
Automated Aggregations: Smart grouping and summarization

🛠️ Processor Code Generation

Automatically generate Sentio processor code for custom blockchain event tracking:

Smart Contract Analysis: Analyze contract ABIs and suggest relevant events
TypeScript Generation: Create complete processor implementations
Test Generation: Include comprehensive test cases
Project Bootstrapping: Set up complete Sentio projects

💬 Interactive Chat Interface

Engage with your data through an intuitive conversation experience:

Context-Aware: Remembers your project context and previous queries
Streaming Responses: Real-time response generation
Resource Attachments: Reference specific tables, metrics, or files
Session Management: Persistent conversations with history

Supported Use Cases

Data Exploration

"Show me the top 10 tokens by trading volume on Ethereum last week"

"What's the average gas price trend over the past month?"

"Find all failed transactions for contract 0x..."

Metrics & KPIs

"Create a daily active users metric for my DeFi protocol"

"Track TVL changes over time with weekly aggregation"

"Show me revenue metrics broken down by token type"

Custom Development

"Generate a processor to track Uniswap V3 swap events"

"Create monitoring for NFT minting on this contract"

"Build analytics for my token's transfer patterns"

Architecture

The AI system uses a multi-service architecture:

AI Service (Go): Session management, authentication, routing
Python AI Server: LLM integration, specialized agents
Frontend Components (React): Chat interface, visualization
Multiple LLM Support: OpenAI GPT-4, Anthropic Claude

Getting Started

Access the AI Chat: Open any project dashboard and click the AI chat panel
Set Context: The AI automatically understands your current project
Ask Questions: Use natural language to query your data
Refine Results: Iterate on queries and visualizations
Save Insights: Export or save generated queries and charts