iTranslated by AI
Snowflake Cortex Search Management UI Has Become More Convenient
Introduction
As we enter 2025, Snowflake's AI capabilities (Cortex AI) are becoming increasingly robust. In particular, Snowflake AI & ML Studio is frequently updated, enabling many AI/ML features to be operated through an intuitive UI. This embodies Snowflake's commitment to ease of use and simplicity.
This time, I'd like to introduce the significantly improved management UI for Cortex Search from Snowflake AI & ML Studio. Previously, managing Cortex Search services relied heavily on SQL commands, but now comprehensive GUI-based management is finally possible. You can check the service list, modify settings, preview data, and even use the playground feature—all within Snowsight.
This allows data engineers and AI application developers alike to leverage Cortex Search more efficiently. I'll explain the value and usage of these new management UIs in detail, accompanied by actual screen captures.
What is Cortex Search?
Let me briefly explain Cortex Search again. Cortex Search is a fully managed enterprise search service provided by Snowflake.
Key Features
- Hybrid Search: High-precision search combining vector search (semantic similarity) and keyword search (lexical similarity)
- Automatic Embedding Generation: Automatically and periodically vectorizes text data to enable semantic search
- Fully Managed: No infrastructure management or tuning required; ready to use immediately
- Snowflake Native: Fully integrated with data governance and security features
For more details on Cortex Search and its applications, please refer to my previous articles "Try RAG Chat App with Snowflake Cortex Search" and "Enhance RAG Chat App with Snowflake Cortex Search Boosts & Decays".
Managing Cortex Search in Snowflake AI & ML Studio
Snowflake AI & ML Studio is provided as Snowflake's GUI-based AI/ML environment, and the ability to create Cortex Search services has been available for some time. With recent updates, the management functionality for Cortex Search has been greatly enhanced, allowing almost all tasks required for creating and operating Cortex Search to be completed within Snowflake AI & ML Studio.
Accessing AI & ML Studio
First, log in to Snowsight, then click 'AI & ML' in the left pane, followed by 'Studio'.
Main screen of AI & ML Studio
Within the Studio, you can manage various AI/ML features such as Cortex Search, Cortex Analyst, LLM Playground, and ML function creation via a web-based UI.
Within the Studio, various AI/ML features such as Cortex Search, Cortex Analyst, LLM Playground, and ML function creation can be managed via a web-based UI.
The New Cortex Search Management UI
1. Cortex Search Service List Screen
When you select Cortex Search from AI & ML Studio, all existing Cortex Search services are displayed in a list.
Cortex Search service list screen
Actions that can be performed on a Cortex Search service
Information visible in the list:
- Service name and status
- Database and schema
- Index creation status
- Index refresh frequency
Actions that can be performed:
- Create a new service ('Create' button)
- View details of an existing service
- Try searches in the Playground
- Manually refresh the index
- Change the index refresh frequency or warehouse
- Pause/resume index creation
- Pause/resume the service
- Delete the service
Information that previously could only be confirmed with SQL commands like SHOW CORTEX SEARCH SERVICES; is now displayed in a visually clear format. This is especially useful because Cortex Search consumes computing resources for index creation, helping to avoid situations like "I accidentally kept updating a Cortex Search service that I wasn't using!"
2. Service Detail Screen and Data Preview
When you select a specific Cortex Search service, you can view detailed configuration information, actual data previews, and Cortex Search cost information.
Service detail screen
Data preview screen
Cost confirmation screen
Detailed information displayed:
- Details of search columns
- Details of attribute columns
- Last updated timestamp
- Warehouse being used
- Embedding model being used
- Service query URL and sample code for calling via Python/API
- Display of actual data previews
- Cortex Search cost information
A particularly noteworthy point is that you can actually view the vector data. While this is often hidden in many managed services, being able to check the actual embedded vector data provides very useful debugging information for AI engineers.
3. Playground Feature
Another welcome feature is the Playground. This allows you to actually try searches.
Cortex Search Playground screen
What you can do in the Playground:
- Execute search queries in natural language
- View search result chunks
- Filter by attributes
- Validate the effects of Boosts & Decays
This enables developers to verify search quality before implementation.
Business Value and Impact
Dramatic Improvement in Development and Operational Efficiency
| Traditional Challenges | Solutions with the New Management UI |
|---|---|
| Requires memorizing and executing SQL commands | Intuitive GUI operations |
| Difficult to grasp service status | Real-time status display and dashboard |
| Time-consuming verification of search quality | Instant testing in the Playground |
| Difficult to obtain debugging information | Visualization of vector data |
| Risk of configuration errors | Guided setup via GUI |
Improved ROI
Reduced Development Time:
- Reduced service management time
- Reduced testing and verification time for search quality
Optimized Operational Costs:
- Real-time monitoring from the GUI enables appropriate resource management
- Pre-testing in the Playground reduces trial and error in production environments
- Reduced re-creation costs due to configuration errors
Democratization of AI Application Development
Traditionally, managing Cortex Search required specialized SQL knowledge, but the GUI has made it much more accessible. This enables:
- Business users to verify search quality and perform basic configuration changes
- Application developers to more easily integrate enterprise search and RAG chat applications
Conclusion
The revamp of the Cortex Search management UI in Snowflake AI & ML Studio has significantly lowered the barrier to leveraging enterprise search and developing RAG chat applications. Operations that previously required specialized SQL knowledge can now be performed through the GUI, and the Playground feature makes it easy to verify search quality.
The following points are particularly valuable:
- Significantly improved development efficiency: Intuitive GUI operations drastically reduce development time
- Improved quality: Pre-testing in the Playground enhances search quality in production environments
- Reduced learning curve: New engineers can start using Cortex Search in a short time
- Strengthened governance: Centralized management within the Snowflake environment maintains security and governance
- Enhanced visibility: Detailed information including vector data is visualized
If you are considering developing RAG applications or building enterprise search systems, please try out this new management UI. Snowflake's AI capabilities are evolving as a true "AI platform for enterprise data", continuously improving in both usability and functionality.
Promotions
Delivered a Virtual Hands-On at SNOWFLAKE DISCOVER!
At the large-scale webinar for engineers, "SNOWFLAKE DISCOVER Vol. 2" held on July 8-9, 2025, I delivered a practical virtual hands-on session titled "Next-Generation VoC (Voice of Customer) Analysis with Snowflake Cortex AI." This session allows you to experience how to analyze customer voices, relevant to many industries, using Snowflake's latest features. If you're looking for insights on analyzing unstructured data, we hope you'll watch it!
You can register via the link below and watch it on-demand immediately.
Presented at SNOWFLAKE DISCOVER!
At the large-scale webinar for engineers, "SNOWFLAKE DISCOVER" held on April 24-25, 2025, I spoke in the very first session titled "Starting from Zero with Snowflake: Building a Modern Data & AI Platform." I explained everything from the basics of Snowflake to the latest developments in as clear a way as possible, so we hope you'll use it for catching up!
You can register via the link below and watch it on-demand immediately.
Presented at a Webinar by Generative AI Conf!
In a webinar themed "Platforms Supporting the Generative AI Era," I participated as a Snowflake employee alongside NVIDIA and my former company AWS, giving a lightning talk on Data x AI. Below is the video archive. We hope you'll watch it!
Sharing Snowflake What's New Updates on X
I'm sharing Snowflake What's New updates on X. Feel free to follow along!
Japanese Version
Snowflake What's New Bot (Japanese Version)
English Version
Snowflake What's New Bot (English Version)
Revision History
(20250705) Initial post
(20250927) Updated promotions section
Discussion