Join us for an interactive live code-along session where we'll take a deep dive into building a complex travel assistant using Rasa's CALM paradigm. This session will guide you through the process of creating a powerful AI assistant capable of handling tasks like flight bookings, hotel reservations, and more—all while optimizing for cost efficiency and reliability.
You'll be coding along with Alan Nichol, cofounder and CTO of Rasa, and Hugo Bowne-Anderson, independent data and AI scientist.
What You’ll Learn
- Building a Travel Assistant: Step-by-step guidance on implementing a sophisticated travel assistant using CALM, with features like flight booking, hotel reservations, and car rentals.
- Hands-On Coding: Participate in live coding exercises that bring these concepts to life within a GitHub Codespace.
- Optimizing Performance: Learn how to evaluate the assistant’s performance on metrics such as token usage, latency, and accuracy, and visualize the results.
- Domain-agnostic techniques: Although we'll be building a travel assistant, know that all the techniques, concepts, and tools we'll cover are totally domain-agnostic!
Who Should Attend:
- AI developers and practitioners looking to build complex, cost-efficient AI assistants.
- Enterprise users interested in leveraging AI for travel-related tasks and beyond.
- Anyone eager to dive into advanced AI assistant development techniques.
What You’ll Need
- Just a GitHub account! We’ll take care of everything else to get you up and running.
Engage and Learn
This is a hands-on session designed to be interactive. Come prepared to code, ask questions, and enhance your skills in building AI assistants.
0:00 Introduction and Welcome
2:06 Live Coding Session Overview
3:41 Enhancing LLMs with Business Logic
5:20 Calm: Separating Business Logic from Conversation
8:57 The Power of Structured Conversations
13:35 Demo: Building a Travel Assistant with CALM
23:48 Setting Up the Development Environment
28:11 Understanding Rasa Actions and Components
30:38 Interacting with the Travel Assistant
35:31 Evaluating Conversational AI Systems
39:43 Comparing CALM, LangGraph, and NLU Approaches
45:16 Deep Dive: CALM's Flow Structure and Logic
49:12 Customizing Responses and Managing Risk
52:51 Configuring Language Models in CALM
54:30 Upcoming Webinar: Advanced CALM Techniques
55:33 Q&A: Optimizing Model Selection for Different Tasks
Want to build your own chatbot? [ Ссылка ]
Learn more about Rasa: [ Ссылка ]
Rasa documentation: [ Ссылка ]
Join the Rasa Community: [ Ссылка ]
Twitter: [ Ссылка ]
Facebook: [ Ссылка ]
Linkedin: [ Ссылка ]
