Transform Salesforce data interaction with AI-infused intelligent query assistance, visual schema exploration, and advanced SOQL development.
Click here to view full-resolution features showcase
Experience the complete DataPilot AI platform in action
- Advanced Salesforce Data Platform
- Core Features
- Technical Architecture
- Complete Documentation Suite
- Get Started with DataPilot
- Deployment & Development Documentation
- Roadmap
- Contributing
- License
Click on any feature below to explore detailed documentation and capabilities
INTELLIGENT NATURAL LANGUAGE TO SOQL CONVERSION
DataPilot's flagship AI feature - Advanced natural language processing for Salesforce queries
- Custom ReAct Agent: LangGraph-based agent for Salesforce operations
- Real-time Streaming: Live responses with intelligent chunked delivery
- Persistent Conversation Memory: Conversation state management with MongoDB
- 4 Specialized Salesforce Tools: Object search, metadata, relationships, and query execution
- Data Redaction Strategy: Optimized LLM token usage with complete data security
- Multi-turn Conversations: Context-aware follow-up interactions
Key Capabilities:
- Converts natural language to complex SOQL queries
- Maintains conversation context across interactions
- Provides intelligent query suggestions
- Streams responses in real-time for better user experience
Supported Use Cases:
- Object Discovery: "Find all objects related to customer data"
- Field Exploration: "What fields are available on the Account object?"
- Relationship Mapping: "How are Account and Contact objects related?"
- Data Queries: "Show me all opportunities closing this month"
- Complex Queries: "Get accounts with their related contacts and opportunities"
- Field Details: "What are the picklist values for the Status field on Case?"
- Multi-turn Conversations: Build complex queries through iterative refinement
VISUAL SCHEMA EXPLORATION FOR SALESFORCE METADATA
- Hierarchical Tree Navigation: Expandable tree view of Salesforce objects and fields
- Visual Canvas Interface: Interactive drag-and-drop schema visualization
- Smart Relationship Mapping: Visual relationship intelligence and navigation
- Performance-Optimized Caching: Intelligent caching for instant access
- Drag-and-Drop Integration: Seamless integration with query editor
INTELLIGENT SOQL DEVELOPMENT ENVIRONMENT
- 6-Level Context-Aware Autocomplete: SObject, field, relationship, keyword, function, nesting
- Professional Code Formatting: Advanced query formatting
- Bidirectional Integration: Seamless schema canvas integration
- Multi-Hierarchy Results: Advanced result visualization with 4-level nesting
- Real-time Validation: Instant error detection and correction
- Inline Record Editing: Edit field values directly in the results grid and save changes back to Salesforce
APEX DEVELOPMENT ENVIRONMENT
- Anonymous Apex Execution: Run Apex code against your connected org and view debug logs
- Saved Apex Snippets: Create, edit, and re-run saved Apex code
- Apex Class & Trigger Browser: Load and inspect existing classes/triggers, compile packages and triggers
- Test Execution: Run Apex tests and view results, including compile-and-test workflows
Note on data writes: DataPilot can already modify org data and metadata through inline record editing and Apex execution (both are direct, deliberate user actions). The AI Query Assistant itself is currently read-only — its tools only search, describe, and query Salesforce data; it does not yet perform inserts/updates/deletes on your behalf. See the Roadmap for planned AI-driven data updates.
graph LR
A[Natural Language Input] --> B[AI Agent Processing]
B --> C[Salesforce Tools Execution]
C --> D[Intelligent Data Processing]
D --> E[Context-Aware Response]
E --> F[Real-time Streaming Output]
F --> G[Conversation Learning]
G --> B
- React + TypeScript: Modern, type-safe development
- Vite Build System: Lightning-fast development
- Mantine UI: Professional, accessible components
- Real-time WebSockets: Live streaming communication
- Python + FastAPI: High-performance API framework
- LangGraph Integration: Advanced AI workflow orchestration
- MongoDB: Scalable data persistence
- OpenAI Integration: Cutting-edge language models
- AI-Powered Query Assistant - Natural language to SOQL conversion
- Interactive Schema Explorer - Visual metadata exploration
- Advanced SOQL Query Editor - Intelligent code editor
- Apex Code Management - Apex development environment
- Clone the repository and follow the setup instructions
- Backend Setup: Follow the Backend README for Python/FastAPI setup
- Frontend Setup: Follow the Frontend README for React/TypeScript setup
- Docker Deployment: Use the Docker README for containerized deployment
- Configure your Salesforce connection
- Start using the AI-powered query assistant
- Integration Setup: API and external system connections
- Monitoring: Usage monitoring
The Docker Compose setup seeds default MongoDB credentials for local development. Change these before exposing DataPilot beyond your own machine:
| Variable | Where | Default | Notes |
|---|---|---|---|
MONGO_USER / MONGO_PASS |
docker/docker-compose.yml, docker/environment-configs/backend.env |
datapilot / datapilot123 |
Change immediately for any non-local deployment |
LLM_API_KEY |
docker/environment-configs/backend.env |
placeholder | Required for the AI Query Assistant; set your own OpenAI/Groq/Ollama key |
LLM_PROVIDER |
docker/environment-configs/backend.env |
openai |
openai, groq, or ollama |
LANGFUSE_* |
docker/environment-configs/backend.env |
placeholder / disabled | Optional; enables LLM call tracing |
VITE_API_BASE_URL |
docker/environment-configs/dashboard.env |
http://localhost:8001 |
Must be changed to your server's address for any deployment the browser doesn't run on the same machine as the backend — localhost in a served frontend resolves to the viewer's machine, not your server |
CORS_ALLOW_ORIGINS |
docker/environment-configs/backend.env |
["*"] |
Restrict for production use |
See Docker README for the full configuration walkthrough.
- Docker README - Complete docker deployment guide
- Backend README - Complete backend development guide
- Frontend README - Complete frontend development guide
- Salesforce Web-based Authentication: Native Salesforce OAuth integration
- Multi-tenant Support: Support for multiple Salesforce orgs
- Record Insert & Delete: The UI currently supports inline record updates; insert and delete operations through the UI are planned
- AI-Powered Data Updates: Natural language data modification through the AI agent (the AI agent is currently read-only)
- Data Validation: Simple data validation and error handling
Contributions are welcome! See CONTRIBUTING.md for setup instructions and guidelines. For security issues, see SECURITY.md.
This project is licensed under the MIT License - see the LICENSE file for details.
ADVANCED SALESFORCE DATA PLATFORM
Transform your Salesforce experience with intelligent query assistance, visual schema exploration, and advanced SOQL development.