Introduced at Subsequent 25, the Looker Conversational Analytics API serves because the agentic backend for Looker AI. It solutions questions utilizing a reasoning agent that makes use of a number of instruments to reply analytical questions. It additionally makes use of dialog historical past to reply multi-turn questions and allow extra environment friendly Looker queries, together with the power to open them within the Discover UI.
Looker’s AI structure is designed for accuracy and high quality, taking a multi-pronged method to gen AI high quality:
-
Agentic reasoning
-
A semantic layer basis
-
A dynamic data graph that gives context for Retrieval Augmented Technology (RAG)
-
Tremendous-tuned fashions for SQL and Python era
This sturdy structure permits Looker to maneuver past merely answering “What?” inquiries to addressing extra advanced queries like “How does this examine?” “Why?” “What is going to occur?” and finally, “What ought to we do?”
Looker’s AI and BI roadmap
With Looker, we’re dedicated to converging AI and BI, and are engaged on quite a lot of new choices together with:
-
Code Interpreter for Conversational Analytics makes superior analytics straightforward, enabling enterprise customers to carry out advanced duties like forecasting and anomaly detection utilizing pure language, without having in-depth Python experience. You possibly can be taught extra about this new functionality and enroll right here for the Preview.
-
The Conversational Analytics API lets builders combine Conversational Analytics throughout a number of experiences, together with buyer purposes, chat apps, Agentspace, and BigQuery. Join right here for preview entry to the Conversational Analytics API.
-
Centralize and share your Looker brokers with Agentspace, which affords centralized entry, quicker deployment, enhanced staff collaboration, and safe governance.
-
Automated semantic mannequin era with Gemini helps democratize LookML creation, increase developer productiveness, and unlock information insights with multi-modal inputs. Gemini leverages numerous enter sorts like pure language descriptions, SQL queries, and database schemas.
Embracing BI’s AI-powered future
Gemini in Looker is a big milestone within the AI/BI revolution. By integrating the ability of Google’s Gemini fashions with Looker’s sturdy information modeling and analytics capabilities, organizations can empower their analysts, improve the productiveness of their enterprise customers, and unlock deeper, extra actionable insights from their information. Gemini in Looker is remodeling how we perceive and leverage information to make smarter, extra knowledgeable choices. The journey from asking “What?” to confidently figuring out “What subsequent?” is now inside attain, powered by Gemini in Looker. Be taught extra at https://cloud.google.com/looker, or click on right here to be taught extra about Gemini in Looker and the best way to allow it to your Looker deployment. You can even select to allow Trusted Tester options to realize entry to early options in growth.
