Snowflake, a leading data cloud company, has announced the launch of Cortex Analyst, an agentic AI system that simplifies self-service analytics through powerful language models. The fully managed service provides businesses with a conversational interface, allowing users to ask questions in plain English and receive accurate insights.
The current ecosystem of analytics heavily relies on business intelligence (BI) dashboards, which can sometimes be rigid and limit users from delving deeper into specific metrics. Snowflake’s Cortex Analyst aims to bridge this gap by leveraging large language models (LLMs) to deliver accurate and context-aware answers to user queries. This eliminates the need for analysts and accelerates the decision-making process.
Baris Gultekin, Snowflake’s head of AI, emphasizes the importance of accuracy when asking business questions. Snowflake’s Cortex Analyst achieves an accuracy of about 90%, surpassing existing LLM-powered text-to-SQL offerings. By utilizing a series of LLMs working together, Cortex Analyst significantly improves the accuracy of analytical insights.
To ensure accurate responses, Cortex Analyst requires customers to provide semantic descriptions of their data assets. This allows the LLM agents to understand the complete schema of the user’s data structure and capture the intent of the question. Snowflake’s customers can specify metrics and their meanings in the semantic descriptions, enabling the system to provide more precise answers.
Currently available as a REST API, Cortex Analyst can be integrated into any application, providing developers with flexibility in how they incorporate the service. Snowflake also offers the option of using Streamlit to build dedicated apps using Cortex Analyst as the central engine.
During the private preview, Cortex Analyst was deployed by enterprises like Bayer to accelerate analytical workflows. With the public preview, Snowflake expects more companies to adopt Cortex Analyst, allowing them to harness the power of LLMs for analytics without the implementation hassle and cost overhead.
Snowflake has plans to add more features to Cortex Analyst, including support for multi-turn conversations and complex tables and schemas. This continuous development aims to provide an interactive and comprehensive experience for users.
In conclusion, Snowflake’s Cortex Analyst offers an innovative solution to simplify analytics and improve decision-making through powerful language models. With its high accuracy and user-friendly interface, Cortex Analyst has the potential to revolutionize the way businesses interact with their data, enabling faster insights and more informed decisions.