Building Effective Conversational AI Systems with Alex Weidauer of Rasa
The real goal of Conversational AI is not to make users believe they are speaking with a human. The real goal is to enable robots to solve issues faster and more effectively, using conversation. Chatbot technology was among the first AI use cases that received early hype. Many early attempts to build failed due to an inability to truly solve issues or due to poor user experience. As the hype has died down, companies are matching the demand and usefulness of Conversational AI with a more educated approach to deploying robots that can communicate. Behind this push for innovation sits Rasa, a high growth startup providing tools for companies and developers to build and deploy Conversational AI

In this episode, I sit with Alex Weidauer, Co-Founder and CEO of Rasa. We discuss topics such as the many pivots his company made to arrive at an AI solution, how Rasa enable developers to build better assistants, how people prefer to communicate with robots, the best approach to training robots on conversation data, and much much more.

If your company is looking to scale it’s AI initiatives, head over to Tesoro AI ( – experts in AI strategy, staff augmentation, and AI product development.
Show Notes with Timestamps (VIDEO):
1:56 The many iterations of Rasa as a startup (before joining the chatbot world)
4:00 The Genesis of Rasa as a Conversational AI infrastructure company
6:03 What is Rasa and how is it used? | Enabling developers to build better assistants
9:13 A historical perspective on when building Conversation AI works and when it fails?
13:47 How do people want to communicate with a robot?
15:08 Most valuable use cases for Conversational AI
17:49 3 Components of Rasa platform | Understanding text, dialogue management, and response
20:00 How Rasa approaches conversation data when building and training system
23:20 How Rasa is using TensorFlow and Googles deep learning framework
25:23 Biggest challenge in building and deploying AI
27:12 Who are Rasa’s users and how technical does the user need to be?
28:45 The importance of a specialist in conversational design
30:22 Alex’s favorite tech tool: Todoist
31:00 Who inspires Alex: Ross Mason of MuleSoft


Share on facebook
Share on twitter
Share on linkedin
Diego Oppenheimer - Headshot

Previous Episode: Diego Oppenheimer, Co-Founder & CEO of Algorithmia

Deploying AI Models at Scale