Data Science Overview

Rierino provides low/no-code capabilities for customizing and managing data science models.

What the Data Science app does

The Data Science app is where you configure and govern ML and GenAI capabilities in a reusable way. You define model parameters, execution settings, and supporting assets so they can be invoked from real-time APIs (runners + sagas) or from batch jobs (Python processes, schedulers, and offline pipelines).

Data Science App

Core Data Science capability areas

Data Science capabilities are organized into these areas:

• ML Models: Configure traditional machine learning models, their inputs, and runtime settings in ML Models. Use this when you want consistent inference behavior across environments.

• GenAI Models: Manage LLM providers, credentials, and agent-facing configuration in GenAI Models. This is the starting point for agent APIs and GenAI troubleshooting.

• MCP Servers: Expose existing platform capabilities over MCP by configuring MCP Servers. Use this when you want tools and microservices to be consumable by MCP clients.

• Complex Event Processing: Define real-time stream logic and windowed aggregations in Complex Event Processing. This is typically used for CEP and near-real-time enrichment pipelines.

• Data Visualizations: Publish embedded dashboards and reporting views through Data Visualizations. Use this for observability of outcomes and business-facing reporting.

How Data Science components work together

In practice, you configure assets here and invoke them elsewhere. A saga step might call an ML/GenAI handler for inference, a batch task might train or backfill features, and CEP flows might continuously compute derived signals. Keeping those definitions centralized makes changes traceable and easier to roll out safely.