Build conversational AI experiences powered by LLMs with Vertex AI Conversation and Dialogflow CX

Conversational interfaces are among the most widely-applicable generative AI use cases. Whether you’re setting up chatbots in just a few steps or creating deep customizations and conversation flows, Google Cloud Dialogflow CX - now with Generative AI features - and Vertex AI Conversation provide tools that make it easy for your business to create personalized, richer experiences for both employees and customers.

In this session, tune into Google Cloud Champion Innovator and Google Developer Expert in Machine Learning, Xavier Portilla Edo, and Developer Advocate, Alessia Sacchi, to learn:
- The latest Generative AI features in Vertex AI Conversation and Dialogflow CX
- How to combine traditional agent design techniques and best practices with Google's latest generative large language models (LLMs) to create complex conversational applications that are prepared to handle the many different ways that users might interact with it
- Examples of tasks that Conversational AI, Search and LLMs can solve for, including demos

04:05 High level overview of natural language understanding for Conversational AI
05:24 Generative AI applied to Conversational AI
11:18 Common user needs, use cases, and capabilities
14:02 Solution: Data store agent (uses retrieval-augmented generation to generate responses grounded to the content of your knowledge repository)
19:06 Solution: Generative fallback (use LLMs to generate virtual agent responses to handle unexpected intents and invalid parameter values)
22:21 Solution: Generators (allows you to make a call to an LLM natively from Dialogflow CX to do anything you would normally ask an LLM to do)
26:21 Live demo
37:25 Best practices when using Generative AI with Conversational AI
48:10 Resources and what's next
50:10 Q&A

Codelabs:

[rboxpdw]

How frequent the datastores can be refreshed. Is it possible to store live flight information in datastore and seat availability?

[ranadeepghosh]

Very helpful! Is it possible to get the deck which was being shared for self notes?

[LazerMarsupial]

How does this differ from Generative Playbooks?

[ShaneCrenshaw]

I wrote a custom application that used generative AI for intent recognition, and then subsequent prompts to gather the required "slots". Judging by the "will dialogflow entities and intents be replaced by gen ai functionality", does that mean this is not able to be done in Dialogflow CX? Thanks!

[nicfeller]

The eligibility quiz this in completely ridiculous... why not just fill out a form???

[naelsonsaraiva]

i am using it now, it's a garbage. I am spent much time with dialogflow CX, langchain be better.