



AWS Comprehend is a natural language processing (NLP) service that uses machine learning to extract insights from text, such as sentiment analysis, entity recognition, key phrases, language detection, and syntax analysis for building intelligent text analytics applications.
Have any questions? We’re here to help You
Makini maintains a comprehensive data model built from analyzing thousands of industrial systems. When data flows through Makini, we automatically transform it from the source system's format into our standardized structure. For example, purchase orders from SAP, NetSuite, and Dynamics all return with consistent field names, data types, and structures. This normalization happens in real-time as data passes through the API. You also have access to raw data if needed for specific use cases. The unified model covers common entities like purchase orders, work orders, inventory items, vendors, and assets, with extensive field coverage across systems.
Data synchronization frequency is configurable based on your requirements. For real-time needs, Makini supports webhook-based synchronization where changes trigger immediate updates. For scheduled syncing, common intervals range from every 15 minutes to daily, depending on data volume and business requirements. The initial sync after connecting a system retrieves historical data based on your configuration—typically 30-90 days of historical records. Subsequent syncs are incremental, retrieving only records created or modified since the last sync. Sync frequency doesn't affect pricing. You can also trigger manual syncs on-demand via API when needed for specific workflows.
Webhooks allow Makini to notify your application of events in real-time. To set up webhooks, configure a webhook URL in your connection settings or during the initial connection flow. Your webhook endpoint must accept POST requests, respond within 10 seconds with a 200 status code, and use HTTPS with a valid SSL certificate. Makini will send webhook payloads to your endpoint when configured events occur, such as sync completion, connection status changes, or errors requiring attention. We recommend keeping your webhook receiver lightweight—ideally just writing the payload to a queue for asynchronous processing—to avoid timeouts and ensure reliable delivery.
Makini sends webhooks for several event types: sync completion (successful or failed), connection authentication required (when credentials need renewal), connection status changes (online/offline), and system errors requiring attention. Each webhook payload includes the event type, timestamp, connection ID, and event-specific details like error messages or affected entities. You can configure which events trigger webhooks on a per-connection basis. For workflow-based integrations using Makini Flows, you can also set up custom webhooks triggered by specific conditions in your business logic, providing granular control over real-time notifications.
