Import to Infino
Create a dataset for cross-source correlations or enrichment. Upload files or load sample datasets for prototyping and experimentation.
Access Import via Data → Connectors in the sidebar (click "Upload files" pill).
File Upload
Upload Interface
- Drag & drop zone - Drop files directly or click to browse
- Multi-file selection - Upload multiple files to create a single merged dataset
- File format validation - JSONL, CSV, and JSON
File Processing
- Schema analysis - Automatic field type detection and preview
- Merge capabilities - Multiple files combined into unified schema
- Error validation - Duplicate file detection and format verification
Import Workflow
You upload files:
- Drag files to upload zone or click "Select Files" button
- System automatically analyzes file schemas and shows preview
- Click "Proceed to Import" to configure dataset settings
- You are automatically redirected to Dataset Detail page upon completion
Sample Datasets
Available Datasets
- Flight Data - Aviation records with departure/arrival information
- E-commerce Data - Transaction and product data
- Multiple categories - Various domain-specific datasets
Dataset Features
- Instant loading - No file upload required
- Pre-configured schemas - Ready for immediate analysis
- Status indicators - Shows "Added" badge for previously imported datasets
Sample Data Workflow
You select datasets:
- Click dataset cards to view details and preview
- Click "Load This Dataset" to import
- You are automatically redirected to Dataset Detail page upon completion
Context Enrichment
After importing data, Fino automatically generates semantic context for your datasets:
- Field Descriptions - AI-generated explanations of what each field contains
- Synonyms - Alternative names to improve natural language query accuracy
- Filters - Pre-defined query conditions
Access context via Data → Sources → FinoDB, select a dataset, then click the Context tab.
For detailed information, see Context Enrichment.
Use Cases for Infino Import
Import data into Infino when:
- Prototyping: Quick experimentation with sample datasets
- Data enrichment: Store enriched results and more context
- Cross-source correlations: Correlate across diverse data sources and formats
- Edge Cases: Handle data that doesn't fit existing schemas
For production workloads, prefer querying data in place via Data → Connectors.
Transforms
Transforms let you materialize repeatable queries into new datasets on a schedule (or on-demand).
- Manage transforms: Navigate to Data → Transforms
- Create from chat: In a Conversation, use Save or Transform (next to Ask / Generate / Raw)
- Ask mode: saves the natural-language prompt as a
finotransform - Generate mode: saves the generated SQL/QueryDSL
- Raw mode: saves the raw SQL/QueryDSL you provided
- Transform: saves + executes, then switches the conversation’s dataset selection to the output dataset
- Ask mode: saves the natural-language prompt as a
Import Jobs
Monitor Jobs
Navigate to Data → Sources and click Jobs button in the header to track import progress:
- View job status (running, completed, failed)
- Monitor real-time processing metrics
- Access job history and error logs
Job Management
- Click job entries for detailed information
- Track records processed, file sizes, processing time
- Filter jobs by status