Thursday, June 13, 2024
HomeMachine LearningDeploy a Microsoft Groups gateway for Amazon Q, your corporation skilled

Deploy a Microsoft Groups gateway for Amazon Q, your corporation skilled

Amazon Q is a brand new generative AI-powered utility that helps customers get work finished. Amazon Q can turn out to be your tailor-made enterprise skilled and allow you to uncover content material, brainstorm concepts, or create summaries utilizing your organization’s information safely and securely. You need to use Amazon Q to have conversations, clear up issues, generate content material, acquire insights, and take motion by connecting to your organization’s data repositories, code, information, and enterprise methods. For extra data, see Introducing Amazon Q, a brand new generative AI-powered assistant (preview).

On this publish, we present you the right way to deliver Amazon Q, your corporation skilled, to customers in Microsoft Groups. (When you use Slack, confer with Deploy a Slack gateway for Amazon Q, your corporation skilled.)

You’ll have the opportunity converse with Amazon Q enterprise skilled utilizing Groups direct messages (DMs) to ask questions and get solutions based mostly on firm information, get assist creating new content material equivalent to e mail drafts, summarize connected information, and carry out duties.

You can even invite Amazon Q enterprise skilled to take part in your Groups channels. In a channel, customers can ask Amazon Q enterprise skilled questions in a brand new message, or tag it in an present thread at any level, to offer further information factors, resolve a debate, or summarize the dialog and seize the subsequent steps.

Resolution overview

Amazon Q enterprise skilled is amazingly highly effective. Take a look at the next demo—seeing is believing!

Within the demo, our Amazon Q enterprise skilled utility is populated with some Wikipedia pages. You’ll be able to populate your Amazon Q enterprise skilled utility with your personal firm’s paperwork and data base articles, so it is going to be capable of reply your particular questions!

Every little thing you want is offered as open supply in our GitHub repo.

On this publish, we stroll you thru the method to deploy Amazon Q enterprise skilled in your AWS account and add it to Microsoft Groups. Once you’re finished, you’ll surprise the way you ever managed with out it!

The next are a few of the issues it may well do:

  • Reply to messages – In DMs, it responds to all messages. In channels, it responds solely to @mentions and responds in a dialog thread.
  • Render solutions containing markdown – This consists of headings, lists, daring, italics, tables, and extra.
  • Observe sentiment – It offers thumbs up and thumbs down buttons to trace person sentiment.
  • Present supply attribution – It offers references and hyperlinks to sources utilized by Amazon Q enterprise skilled.
  • Perceive dialog context – It tracks the dialog and responds based mostly on the context.
  • Keep conscious of a number of customers – When it’s tagged in a thread, it is aware of who stated what, and when, so it may well contribute in context and precisely summarize the thread when requested.
  • Course of connected information – It will probably course of as much as 5 connected information for doc query answering, summaries, and extra.
  • Begin new conversations – You’ll be able to reset and begin new conversations in DM chats by utilizing /new_conversation.

Within the following sections, we present the right way to deploy the mission to your personal AWS account and Groups account, and begin experimenting!


It is advisable to have an AWS account and an AWS Identification and Entry Administration (IAM) position and person with permissions to create and handle the required sources and elements for this utility. When you don’t have an AWS account, see How do I create and activate a brand new Amazon Net Companies account?

You additionally must have an present, working Amazon Q enterprise skilled utility. When you haven’t set one up but, see Creating an Amazon Q utility.

Lastly, you want a Microsoft account and a Microsoft Groups subscription to create and publish the app utilizing the steps outlined on this publish. When you don’t have these, see if your organization can create sandboxes so that you can experiment, or create a brand new account and trial subscription as wanted to finish the steps.

Deploy the answer sources

We’ve offered pre-built AWS CloudFormation templates that deploy all the things you want in your AWS account.

When you’re a developer and also you wish to construct, deploy, or publish the answer from code, confer with the Developer README.

Full the next steps to launch the CloudFormation stack:

  1. Log in to the AWS Administration Console.
  2. Select one of many following Launch Stack buttons to your desired AWS Area to open the AWS CloudFormation console and create a brand new stack.
Area Launch Stack
N. Virginia (us-east-1)
Oregon (us-west-2)
  1. For Stack identify, enter a reputation to your app (for instance, AMAZON-Q-TEAMS-GATEWAY).
  2. For AmazonQAppId, enter your present Amazon Q enterprise skilled utility ID (for instance, 80xxxxx9-7xx3-4xx0-bxx4-5baxxxxx2af5). You’ll be able to copy it from the Amazon Q enterprise skilled console.
  3. For AmazonQRegion, select the Area the place you created your Amazon Q enterprise skilled utility (us-east-1 or us-west-2).
  4. For AmazonQUserId, enter an Amazon Q enterprise skilled person ID e mail deal with (depart clean to make use of a Groups person e mail because the person ID).
  5. For ContextDaysToLive, enter the size of time to maintain dialog metadata cached in Amazon DynamoDB (you may depart this because the default).

When your CloudFormation stack standing is CREATE_COMPLETE, select the Outputs tab, and maintain it open—you’ll want it in later steps.

Register a brand new app within the Microsoft Azure portal

Full the next steps to register a brand new app within the Microsoft Azure portal:

  1. Go to the Azure Portal and log in along with your Microsoft account.
  2. Select New registration.
    1. For Title, present the identify to your app. You’ll be able to maintain issues easy by utilizing the stack identify you used for the CloudFormation stack.
    2. For Who can use this utility or entry this API?, select Accounts on this organizational listing solely (AWS solely – Single tenant).
    3. Select Register.
    4. Word down the Utility (consumer) ID worth and the Listing (tenant) ID from the Overview web page. You’ll want them later when requested for MicrosoftAppId and MicrosoftAppTenantId.
  3. Select Choose API permissions within the navigation pane.
    1. Select Add a permission.
    2. Select Microsoft Graph.
    3. Select Utility permissions.
    4. Choose Consumer.Learn.All.
    5. Choose ChannelMessage.Learn.All.
    6. Choose Group.ReadBasic.All.
    7. Choose Recordsdata.Learn.All.
    8. Select Add permissions. This permission permits the app to learn information in your group’s listing concerning the signed-in person.
    9. Use the choices menu (three dots) on the best to decide on Take away permission.
    10. Take away the unique Consumer.Learn – Delegated permission.
    11. Select Grant admin consent for Default Listing.
  4. Select Certificates & secrets and techniques within the navigation pane.
    1. Select New consumer secret.
    2. For Description, present a worth, equivalent to description of my consumer secret.
    3. Select a worth for Expires. Word that in manufacturing, you’ll must manually rotate your secret earlier than it expires.
    4. Select Add.
    5. Word down the worth to your new secret. You’ll want it later when requested for MicrosoftAppPassword.
  5. Optionally, select Homeowners so as to add any further homeowners for the applying.

Register your new app within the Microsoft Bot Framework

Full the next steps to register your app within the Microsoft Bot Framework:

  1. Go to the Microsoft Bot Framework and log in along with your Microsoft account.
  2. Optionally, you may create and add a customized icon to your new Amazon Q enterprise skilled bot. For instance, we created the next utilizing Amazon Bedrock picture playground.
  1. Enter your most popular show identify, bot deal with, and outline.
  2. For Messaging endpoint, copy and paste the worth of TeamsEventHandlerApiEndpoint out of your stack Outputs tab.
  3. Don’t choose Allow Streaming Endpoint.
  4. For App sort, select Single Tenant.
  5. For Paste your app ID beneath to proceed, enter the MicrosoftAppId worth you famous earlier.
  6. For App Tenant ID, enter the MicrosoftAppTenantId worth you famous earlier.
  7. Depart the opposite values as they’re, comply with the phrases, and select Register.
  8. On the Channels web page, underneath Add a featured channel, select Microsoft Groups.
  9. Select Microsoft Groups Business (commonest), then select Save.
  10. Comply with the Phrases of Service and select Agree.

Configure your secrets and techniques in AWS

Let’s configure your Groups secrets and techniques so as to confirm the signature of every request and publish on behalf of your Amazon Q enterprise skilled bot.

On this instance, we aren’t enabling Groups token rotation. You’ll be able to allow it for a manufacturing app by implementing rotation through AWS Secrets and techniques Supervisor. Create a problem (or, higher but, a pull request) within the GitHub repo if you would like this characteristic added to a future model.

Full the next steps to configure a secret in Secrets and techniques Supervisor:

  1. On the AWS CloudFormation console, navigate to your stack Outputs tab and select the hyperlink for TeamsSecretConsoleUrl to be redirected to the Secrets and techniques Supervisor console.
  2. Select Retrieve secret worth.
  3. Select Edit.
  4. Exchange the values of MicrosoftAppId, MicrosoftAppPassword, and MicrosoftAppTenantId with the values you famous within the earlier steps.

Deploy your app into Microsoft Groups

Full the next steps to deploy the app to Groups:

  1. Go to the Developer Portal for Groups and log in along with your Microsoft Groups person account.
  2. Select Apps within the navigation pane, then select New app.
    1. For Title, enter your bot identify.
    2. Enter a reputation for Full identify and each brief and full descriptions (you should use the bot identify for all of them if you would like, simply don’t depart them empty).
    3. Enter values for Developer data and App URLs. For testing, you may make up values, and URLs like Use actual ones for manufacturing.
    4. For Utility (consumer) ID*, enter the worth of MicrosoftAppId from earlier.
    5. Select Save.
  3. Beneath Branding, you may add AI-generated icons, or completely different icons, or none in any respect, it’s as much as you. The next are some examples:
    1. Colour icon 192×192
    2. Define icon 32×32
  4. Beneath App options, select Bot.
    1. Choose Enter a bot ID, and enter the MicrosoftAppId worth from the sooner steps.
    2. Beneath What can your bot do?, choose Add and obtain information.
    3. Beneath Choose the scopes during which individuals can use this command, choose Private, Group, and Group chat.
    4. Select Save.
  5. Select Publish.
  6. Select Obtain the app bundle to obtain a .zip file to your laptop.
  7. Select Preview in Groups to launch Microsoft Groups (work or faculty) app.
    1. Within the navigation pane, select Apps, then Handle your apps, then Add an app.
    2. Select Add an app to your orgs app catalog, and choose the .zip file you downloaded. This provides the app to Groups.
    3. Choose the cardboard to your new app, select Add, and look ahead to it to finish (10–20 seconds).

Add your bot to a number of groups

Full the next step so as to add your bot to a workforce:

  1. Within the Groups app, choose your workforce and select Handle workforce.
  2. On the Apps tab, select the brand new Amazon Q enterprise skilled app, and select Add.

Now you may check your bot in Microsoft Groups!

Begin utilizing Amazon Q enterprise skilled

Full the next steps to begin utilizing Amazon Q enterprise skilled in Groups:

  1. Open your Groups consumer.
  2. Beneath Apps, add your new Amazon Q enterprise skilled app to a chat.
  3. Optionally, add your Amazon Q enterprise skilled app to a number of Groups channels.
  4. Within the app DM chat, enter Whats up.

You will have now deployed a robust new AI assistant into your sandbox Groups atmosphere.

Play with it, strive all of the options mentioned on this publish, and replica the stuff you noticed within the demo video. Most significantly, you may ask about matters associated to the paperwork that you’ve got ingested into your personal Amazon Q enterprise skilled utility. However don’t cease there. You could find further methods to make it helpful, and whenever you do, tell us by posting a remark.

As soon as you’re satisfied how helpful it’s, discuss to your Groups admins (present them this publish) and work with them to deploy it in your organization’s Groups organizations. Your fellow workers will thanks!

Clear up

Once you’re completed experimenting with this resolution, delete your app in Microsoft Groups, Bot Framework, and Azure portal. Then clear up your AWS sources by opening the AWS CloudFormation console and deleting the AMAZON-Q-TEAMS-GATEWAY stack that you simply deployed. This deletes the sources that you simply created by deploying the answer.


The pattern Amazon Q enterprise skilled Groups utility mentioned on this publish is offered as open supply—you should use it as a place to begin to your personal resolution, and assist us make it higher by contributing again fixes and options through GitHub pull requests. Discover the code, select Watch within the GitHub repo to be notified of recent releases, and test again for the newest updates. We’d additionally love to listen to your options for enhancements and options.

For extra data on Amazon Q enterprise skilled, confer with the Amazon Q (For Enterprise Use) Developer Information.

In regards to the Authors

Gary Benattar is a Senior Software program Improvement Supervisor in AWS HR. Gary began at Amazon in 2012 as an intern, specializing in constructing scalable, real-time outlier detection methods. He labored in Seattle and Luxembourg and is now based mostly in Tel Aviv, Israel, the place he dedicates his time to constructing software program to revolutionize the way forward for Human Assets. He co-founded a startup, Zengo, with a concentrate on making digital wallets safe by way of multi-party computation. He acquired his MSc in Software program Engineering from Sorbonne College in Paris.

Bob Strahan

Bob Strahan is a Principal Options Architect within the AWS Language AI Companies workforce.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments