Amazon Q, a GenAI to Understand AWS (and Your Business Docs)

Amazon Web Services has such a bewildering array of cloud services, the company itself now admits that it needs an AI assistant to get the best use from them.

Tuesday, at the company’s annual Re:Invent user conference in Las Vegas,  AWS introduced Amazon Q, a generative AI-powered assistant designed to make sense of the palette of AWS tools available, and even help the business users understand the company data on hand.

Amazon Q, initially in @awscloud’s Cloud Catalyst, is a new generative AI-based chat to help devs get started on building projects within AWS, with step-by-step instructs, debugging, test automation, connects to business apps, your specific domain logic, etc.. #awsreinvent2023 pic.twitter.com/AiTJsPx0B5

— Joab Jackson (@Joab_Jackson) November 28, 2023

Amazon Q provides AWS Expertise

In many cases, Q could help potential users of AWS in a major way, noted Ben Schreiner, AWS head of business innovation, in an interview with TNS.

“AWS is built for builders built by engineers, for engineers, and it is great. It can be intimidating to those that maybe don’t have that background, or, don’t have developers at their disposal,” Schreiner said.

Scheiner’s office helps small and medium businesses (SMBs) get the most from AWS, through the use of solutions libraries and other tools. Unlike enterprises or web-scale companies, most smaller organizations do not have a deep bench of developers or operations help. So a tool like Q could help them get to the next level of cloud usage, Schreiner said.

“The ‘how’ is the hard part,” Schreiner said. With Q, “I don’t have to know how to do it, I just have to know how to ask the question.”

Developers and Analysts

AWS expects two sets of users for Q: developers and analysts.

For developers, Q offers a conversational interface from the IDE that can step users through all the steps to spin up a service. The newbie can ask a questions from the general (“What are AWS serverless services to build serverless APIs?”) to the highly specific (“I’m planning to create serverless APIs with 100k requests/day. Each request needs to look up into the database. What are the best services for this workload?” ):

👁 Image

You can also ask it to suggest the best resource for a particular workload, such as what the best instance is to run a web server. Which is the most cost-effective Gravitron instance should you use for training your models? Q does the calculations.

Amazon Q can also help debug AWS services. Why doesn’t the AWS Lambda function interact with an Amazon DynamoDB table? Q will do some rooting around its extensive knowledge base of AWS services…

👁 Image

This service, now available in preview, is available Amazon CodeCatalyst and as a plug-in for Visual Studio and will soon also be extended to various IDEs.

Amazon Q for Business Users

Q is also available for QuickSight, Amazon’s tool for business analysis, offering to help out with some of the more routine tasks of the business manager.

For instance, it can create projects from a set of data, write executive summaries and even, in AWS words, build a narrative.

Using natural language on the Amazon QuickSight dashboard, a user can ask to build a story around a set of data. Q will extract data insights and statistics from selected visuals, then summarize what the data may mean for the business, and even suggest ideas for specific goals.

👁 Image

Pricing for Q starts at US$20/month for the business package and $25/month for the developer edition.

.@awscloud Redshift query editor now can turn natural language into SQL, thx to #AmazonQs understanding the data warehouse tables. “If you need help creating custom SQL, you can turn your natural language prompts into customized recommendations.”-@SwamiSivasubram #awsreinvent2023 pic.twitter.com/KF1GqJcz5d

— Joab Jackson (@Joab_Jackson) November 29, 2023