re:Invent Guide: Conversational AI


Machine Learning Hero

This guide is for re:Invent attendees who are as excited about conversing with machines as I am, who want to be inspired by what is possible in this space, and who also want to deliver amazing experiences to real users while following good practices.

I am a passionate advocate of cloud-native architecture with a keen interest in how conversational interfaces, machine learning, and artificial intelligence (AI) services can be applied within an AWS serverless SaaS architecture to deliver amazing new capabilities. As co-founder and head of cloud engineering and AI at Workgrid Software, I have been applying these technologies and techniques in the enterprise environment to deliver an award-winning workplace digital assistant.

Conversational design is critical to delivering a successful experience, so I have highlighted several sessions that will put you on the right path.

You can’t talk to a document, picture, or an audio file, yet most data is stored in this unstructured way, so I’m excited about how advances in state-of-the-art NLP and computer vision can help with content extraction and information retrieval, as this will unleash new capabilities that can help deliver solutions for real user needs. I have selected sessions that show you how.

Once you have been inspired and have built some cool stuff, the next critical step is getting it into the hands of users with quality engineering practices. Reducing time to value will be key to your company’s success, so getting a repeatable pathway to production will help power your success.

Get Inspired and Get Building

How developers can build natural, extensible voice conversations ›

Exploring rich conversational experiences is a great way to get your creative juices flowing. This session is a great introduction to what is possible with multi-turn conversations.


Alexa, what can I do now? ›

It can be tough keeping up with the pace of innovation in the conversational AI space. This session helps you understand current capabilities in Alexa and future trends that will inspire you.


Build and deploy your own two-way text chatbot ›

Nothing beats getting your hands on an opportunity to build out capabilities for yourself. This session helps you build out some awesome AI chatbot capabilities.


Build a voice-based chatbot for your Amazon Connect contact center ›

Building out a functioning call center and a voice-enabled chatbot in just over two hours—sign me up! We truly are living in the future. This is a game changer for many enterprises.


Build a hands-free chatbot with AWS services ›

Building an interactive hands-free chatbot experience with voice-driven dictation and hands-free text input using Amazon Transcribe, Amazon Sumerian, Amazon Polly, and Amazon Lex—boom. Voice truly is the future. Prepare to have your mind blown.


Learn to build an Alexa skill in an hour with Cake Walk ›

Time to value is critical for product success. This workshop introduces you to the skills you will need to build a compelling voice experience from idea to working skill in under 60 minutes.

Designing New Experiences

Learn situational design with Cake Walk ›

Good conversational design is critical for a successful Alexa skill. This workshop helps you learn the design skills you need to build an awesome experience.


Deep dive into building multimodal skills ›

Multimodal visual capabilities can help take your experience to the next level. This session shows you the tools and techniques you need to build beautiful multimodal experiences.


Alexa’s different speaking styles for developers ›

Enriching your conversational experience with speaking styles and other SSML features can help your product stand out from the crowd. This session helps expand your thinking of what is possible with voice experiences.


Build a conversational 3D avatar ›

This session is all about putting a face to your chatbot. Awesome!

Content Extraction

Finding a needle in a haystack: Use AI to transform content management ›

Data is the new oil. This session shows you how to extract data, insights, trends, and value from all of your digital content.


Transform the way you search and interact with enterprise data using AI ›

Searching for answers in the enterprise is hard. This session shows you how you can leverage NLP to build an enterprise search capability to provide you with the answers that your users need.

State of the Art

Alexa Conversations ›

Learn how AI-assisted Alexa Conversations can help reduce the burden of building complex interactions.


Making Alexa sound more human—Amazon’s text-to-speech technology ›

This session covers the history and science behind the magic of Alexa and tips for a voice strategy for your company. I can’t wait for this one. Sign me up.


Building state-of-the-art NLP models with Amazon SageMaker ›

Bidirectional Encoder Representation from Transformers (BERT), Amazon SageMaker, NLP, GluonNLP, topic modeling, coherence, and topic uniqueness. Prepare to have your mind blown with this session.


Amazon SageMaker RL: Solving business problems with RL and bandits ›

This one intrigues me. I have had this thought in the back of my head for a while about the possibility of using reinforcement learning (RL) with chatbots. I’m looking forward to learning how RL could be applied to dialog systems.

Pathway to Production: Beyond the POC

Take an ML model from idea to production using Amazon SageMaker ›

Time to value is critical for success. This workshop gives you the skills needed to get your machine learning (ML) models into the hands of real users.


DevOps your Alexa skill: Why testing and automation matter ›

Fast feedback loops help minimize risk. This session helps you understand how testing, automation, monitoring, and alarms give you the feedback you need to make your skills a success. Your future self will thank you for investing time and effort into quality testing and automation.


Build your first Alexa skill from the ASK CLI ›

Learn about how to programmatically build, test, and deploy your Alexa skill using the new open-source version of the Alexa Skills Kit Command Line Interface (ASK CLI). Start your conversational DevOps journey today!


Based on all the great sessions this year, I am really looking forward to December and to learning more and meeting others who are interested in this space. Conversational interfaces and the underlying technologies are evolving rapidly. Get inspired, and start building!

Explore other AWS Hero Guides ›

Registration now open

Register now