overview

Overview

Skypoint AI Mission: Bring People and Data Together.

Skypoint AI is a comprehensive solution stack designed to empower companies by seamlessly integrating, transforming, and analyzing data from a wide range of sources, enabling users to talk to their data. By leveraging open-source generative AI, enterprise-grade technologies and advanced analytics, Skypoint AI allows businesses to access and visualize key insights from their organizational data, helping them make informed and smart decisions.

Let's delve into the processes that allow Skypoint AI to provide a single source of truth, tracing the data's journey from integration to insights. This journey includes stages of integration, ingestion, transformation, utilization and consumption as illustrated in the figure below:

Skypoint AI Architecture

Explore how Skypoint AI revolutionizes data handling, beginning with seamless data integration, advancing through robust data transformation, harnessing the power of generative AI, and culminating in efficient data delivery and insightful consumption:

Data Integration

The first step in the Skypoint AI platform is the real-time or offline integration of diverse data sources with just a few simple clicks using our tools and application. This includes structured data from databases, SaaS applications, Excel files, and APIs, as well as unstructured data from PDFs, PowerPoint presentations, emails, audio, video, HTML, and documents. With the built-in Skypoint AI Dataflow Connector and over 50+ additional connectors on a scalable ELT platform, Skypoint AI ensures seamless and efficient data collection, thus setting the stage for advanced data processing and analysis. By allowing industry and organizational norms for data labelling, before configuring it, Skypoint AI ensures that the data is properly prepared and ready for ingestion into machine learning models.

Data Transformation

The data undergoes multiple layers of transformation. First, unstructured data is converted into embeddings and stored in vector databases. Then, this data is configured with detailed instructions and data model context. In contrast, structured data gets processed through the medallion architecture, which organizes data into layers based on data quality. Following this, it gets stored in the Databricks Lakehouse, where it is further annotated with metadata to enhance its usability. This process improves clarity and helps identify, link, and merge records across different data sources. This optimization is crucial for understanding the data for both users and machine learning models.

Generative AI & Data Utilization

Once the data is prepared, we proceed to the generative AI phase. The core of this phase is the Skypoint AI orchestra, which quickly generates responses to queries. With access to our databases and multi-LLM architecture, it determines the best data source, whether it's structured, unstructured, or a mix of both and uses sophisticated retrieval techniques to ensure accurate and effective query responses.

Data Delivery & Consumption

Skypoint AI delivers data insights through various tools and platforms, including Skypoint AI Copilot and agents that provide direct, secure access to data. It also offers user connectivity through APIs via the Skypoint Developer Portal, automation tools and supports data visualization through BI tools. This facilitates easy access to data insights for end-users, enabling informed decision-making.

Skypoint AI Solutions

Please find the additional features of the Skypoint AI Platform:

Skypoint AI Copilot: Skypoint AI Copilot helps users with instant insights about their own organization's data through natural language.

Dataflow: Data import, export, connections, types (semantic labels) management, data transformations, and data quality management.

AI Configurator: Leverage AI Configurator to create a personalized Copilot, configure data and fine tune AI settings.

Lakehouse: Integrated data warehouse and data lake using Delta Lake technology as the table format.

Resolve: Machine learning and rule-based customer identity resolution & management.

Profile: 360-degree view of the customer including profile, timelines, and enrichment capabilities.

Predict: Predict customer needs from unified data, ML models built-in, and the ability to integrate custom models.

Activate: Capabilities including audiences, metrics, export, and API developer portal.

Empower: Customer consent management and effortless compliance with global privacy laws (CCPA, GDPR).

Automate: Automate workflows, build power apps, and build dashboards using Skypoint AI’s Power BI, Power App, Power Automate, and Teams connectors.