Skip to main content

Platform Flow

This document explains the operational workflow of an gIQ - the AI-powered geospatial analytics platform designed to streamline data ingestion, model execution, and insight generation. The platform follows a structured workflow of data that guides users from initial data sourcing through to generating and viewing stories and reports

Platform Workflow​

Flowchart shows data entering via File Manager Upload, Data Acquire Module, and Analytical Jobs – Results Upload into Ingestion; users then view/access via Explore, File Manager → Explore, or Workspace; choose between viewing detections, selecting a model, or auto/manual run; all runs flow to Inference & Analytics; results are visualized in Explore/Workspace; outputs are produced as Reports or Stories.

Platform Architecture​

The platform operates through seven distinct operational layers, each serving a specific function in the analytics pipeline:

1. Data Sourcing Layer​

As a user, you can introduce data into the platform through the following parallel entry points:

  • File Manager Upload: Direct file upload functionality for structured datasets.

  • Data Acquire Module: Data acquisition from verified external vendors and partners.

  • Analytical Jobs Results Upload: Results of analytical jobs ingested in the platform.

These entry points operate independently, enabling you bring in data from different sources into the platform.

2. Data Processing & Storage Layer​

All incoming data flows through a centralized Ingestion process. This data consolidation point ensures that regardless of the entry method, your data goes through a three-tiered layer of processing, validation, and storage protocols before it becomes available for analysis.

3. Viewing Data & Accessing Data Layer​

Following ingestion, the platorm branches your data into three distinct pathways:

Explore Module: View your data in an interactive exploration space where you can examine datasets, perform ad-hoc several tasks and actions, and understand your data better through properties, metadata, detections, and image bands (if applicable).

File Manager Module: Organize, catalog, and maintain your data assets within the platform.

Workspace Module: Analyse data and create reports and stories enabling you with a critical decision point.

4. Running AI Models Layer​

The Workspace Module introduces a decision node: "Run Auto AI Model?"

YES pathway: The platform automatically selects and executes appropriate AI models based on data characteristics and analytical objectives. This pathway reduces the technical burden on users by leveraging built-in intelligence.

NO pathway: Enables you to run Manual Run AI Model providing granular control over model selection and parameter configuration.

Simultaneously, the Explore Module supports two user actions:

  • View model run detections: Enables you to examine results from previously executed models

  • Select AI Model: Allows direct model selection from the exploration interface

These pathways converge before proceeding to the analytics phase, ensuring all model executions follow a unified output protocol.

5. Running Analytics Layer​

Model outputs feed into the Inference & Analytics process, where the platform uses built-in intelligence and and prepares analytical insights from model predictions.

6. Visualizing Layer​

Analytical results flow into Visualization of Inference, which transforms quantitative outputs into visual representations, enabling you to view the visualization in the following modules:

  • Explore Module

  • Workspace Module

7. Generating and Viewing Reports & Stories Layer​

The final layer provides three distinct output formats:

Reports Module: View reports that you and/or your team mates in the organization generated.

Reports: Create standardized analytical reports with consistent formatting and metrics.

Stories: Creates narrative-driven presentations that contextualize findings within business scenarios.