Introduction

Kumo is an advanced machine learning platform that leverages graph neural networks (GNNs) to deliver predictive analytics and insights directly from your relational data. Kumo simplifies the machine learning process by eliminating the need for extensive data preparation, allowing you to focus on generating valuable predictions and insights.

Key Features:

• Seamless Integration: Kumo works natively on your Snowflake database to build and deploy machine learning models without the data ever leaving your data warehouse.
• Graph Neural Networks: Utilize GNNs to understand and predict complex relationships within your data, offering superior performance over traditional ML approaches.
• Declarative Querying: Kumo's intuitive predictive query language allows both beginners and advanced users to create and deploy models quickly and efficiently.
• Versatile Applications: Kumo supports a wide range of use cases including personalization, fraud detection, customer retention, anomaly detection, and more.

Snowflake users can now deploy Kumo as a native application running in Snowflake, and can access the Kumo platform through the Snowflake Marketplace and pay for it with Snowflake credits.

A free trial of Kumo in Snowflake is available for qualified Snowflake customers. To get started, sign up here: https://kumo.ai/freetrial

Benefits of using Kumo as a Snowflake Native App

Kumo runs as a Snowflake Native App using Snowpark Container Services, a fully managed container offering designed to facilitate the deployment, management, and scaling of applications within the Snowflake ecosystem. This approach allows Snowflake customers to easily and securely share complex containerized applications, with complete control of their application.

Retain Snowflake’s security, compliance, and governance

Kumo runs inside of your Snowflake environment and inherits all Snowflake security, governance rules, and capabilities (e.g., column and row-level security).  This allows you to seamlessly integrate with existing data management workflows and policies by directly using Snowflake as the data storage and processing platform, leveraging Snowflake's single sign-on (SSO) for authentication.

Control all data inside Snowflake 

When using Kumo in Snowflake, no data leaves your Snowflake environment. Data and artifacts generated by Kumo are stored securely and within your Snowflake environment as new tables of predictions or embeddings. These tables comply with Snowflake client data retention policies and are never accessed or managed by Kumo. Once written, the predictions and embeddings can be called directly by other services for use in business operations.

Manage access: As a Snowflake user, you maintain complete control over Kumo access using Snowflake’s access control mechanisms for the duration of the deployment. You must explicitly grant permission to any objects that require access to Kumo; permissions can be revoked or revised at any point in time.

Build on top of Snowflake

Kumo’s seamless integration with the Snowflake data platform allows you to reuse existing Snowflake workloads like ETL and EDA and code.

Improve Model Performance
Kumo improves accuracy by eliminating hand-crafted feature engineering, instead learning embeddings directly from raw relational data, while providing advanced controls that enable data scientists to add business value at each step. 

Answer business questions in hours - not months
Kumo learns directly from your Snowflake tables, eliminates time-consuming feature engineering, and uses AI to generate a model in hours, not months. Furthermore, Kumo’s automated pipelines keep models fresh so the AI always learns from the latest Snowflake tables.