Creating a Conversational Natural Language Interface for Amazon Athena Queries Using Amazon Nova
In this article, we delve into a groundbreaking solution that utilizes Amazon Bedrock Agents, powered by Amazon Nova Lite, to develop a conversational interface for Athena queries. We use AWS Cost and Usage Reports (AWS CUR) to illustrate this concept, though this solution is versatile enough to adapt for other databases queried via Athena. This innovative approach democratizes data access, enabling users to interact with their data using natural language while maintaining the robust analytical capabilities of Athena.
Enabling Rapid Data-Driven Decisions with Text-to-SQL at Parcel Perform
by Maria Gonzalez, Thomas Brown, and Sarah Green
on 09 JUL 2025
in Amazon Athena, Amazon Bedrock, Amazon Bedrock Knowledge Bases, Business Intelligence, Customer Solutions, Generative AI, Intermediate (200), Supply Chain
At Parcel Perform, the business team frequently requires access to data to address inquiries regarding merchants’ parcel deliveries. For instance, they might ask, “Did we observe an increase in delivery delays last week? If so, which transit facilities were affected, and what was the main cause of the issue?” Previously, the data team had to manually create the query and execute it to retrieve the relevant data. However, with the newly implemented generative AI-driven text-to-SQL functionality within Parcel Perform, the business team can now fulfill their data needs independently using an AI assistant interface. In this article, we discuss how Parcel Perform integrated generative AI, data storage, and data access through AWS services to facilitate timely decision-making. For insights on similar topics, check out this other blog post.
Transforming Life Sciences and Genome Analysis with a Generative AI Prototype Powered by Amazon Bedrock
by Ethan Carter, Rachel Adams, and Noah Wilson
on 28 MAY 2025
in Advanced (300), Amazon API Gateway, Amazon Athena, Amazon Bedrock, Amazon DynamoDB, Amazon Simple Storage Service (S3), AWS Amplify, AWS AppSync, AWS Lambda, Customer Enablement, Customer Solutions, How-To, Technical How-to
This article outlines the deployment of a text-to-SQL pipeline utilizing generative AI models and Amazon Bedrock to facilitate natural language queries against a genomics database. We showcase how to create an AI assistant web interface with AWS Amplify and detail the prompt engineering strategies used to generate SQL queries. Additionally, we provide detailed instructions for deploying the service within your AWS account.
Modernizing Data Science Solutions on AWS for Rocket Companies
by Sophia Turner, Liam Hall, and Mason Scott
on 21 FEB 2025
in Advanced (300), Amazon Athena, Amazon EC2, Amazon Elastic Kubernetes Service, Amazon EMR, Amazon Redshift, Amazon SageMaker, Amazon SageMaker AI, Amazon SageMaker Studio, Amazon Simple Storage Service (S3), Amazon VPC, Architecture, AWS Glue, AWS Lake Formation, AWS Lambda, AWS PrivateLink, Compute, Customer Solutions, Financial Services, Migration Acceleration Program (MAP)
In this article, we explore how we enhanced Rocket Companies’ data science solution on AWS to significantly reduce delivery times from eight weeks to under one hour, improve operational stability, and support by minimizing incident tickets by over 99% within 18 months. This transformation enables 10 million automated data science and AI decisions daily while providing a seamless development experience. For more information on similar innovations, visit this excellent resource.
Extracting Data from Google Cloud Platform BigQuery for No-Code Machine Learning with Amazon SageMaker Canvas
by Olivia Turner and Benjamin Lee
on 28 OCT 2024
in Amazon Athena, Amazon SageMaker, Amazon SageMaker Canvas, Amazon SageMaker Data Wrangler, Intermediate (200), Technical How-to
This article outlines an architectural approach for extracting data from various cloud environments, such as Google Cloud Platform (GCP) BigQuery, without necessitating data movement. This strategy minimizes complexity and overhead associated with transferring data across cloud environments, allowing organizations to harness their diverse data assets for machine learning initiatives. We explain the process of using Amazon Athena Federated Query to pull data from GCP BigQuery, employing Amazon SageMaker Data Wrangler for data preparation, and subsequently utilizing this prepared data to develop ML models within Amazon SageMaker Canvas, a no-code ML interface.
The Evolution of the PGA TOUR’s Generative AI Virtual Assistant from Concept to Prototype
by Alexander Foster, Julia King, and Samuel Turner
on 14 MAR 2024
in Amazon Athena, Amazon Bedrock, Amazon Kendra, Customer Solutions, Experience-Based Acceleration, Generative AI, Technical How-to, Thought Leadership
This guest post, co-authored with Scott Gutterman from PGA TOUR, illustrates how generative artificial intelligence opens new avenues for intelligent systems. Recent advancements in generative AI-based large language models (LLMs) have made them applicable across a wide range of information retrieval applications. Given the diverse data sources, LLMs provide tools for better insights.
Building a Comprehensive Text-to-SQL Solution for Complex Queries and Diverse Data Sources
by Jacob Miller and Emma Wilson
on 28 FEB 2024
in Amazon Athena, Amazon Bedrock, Artificial Intelligence, AWS Glue, Expert (400), Generative AI, Technical How-to, Thought Leadership
Structured Query Language (SQL) is an intricate language that necessitates a solid understanding of databases and metadata. Today, generative AI empowers individuals lacking SQL proficiency. This generative AI task, known as text-to-SQL, generates SQL queries through natural language processing (NLP) and converts text into semantically accurate SQL. The solution presented in this article aims to bridge the gap between complex querying and user accessibility, enhancing the overall experience of data interaction. For authoritative insights, refer to this link.
Leave a Reply