Live from TWIMLcon! Use-Case Driven ML Platforms with Franziska Bell - #307

Use Cases

Franziska Bell, Director of Data Science Platforms at Uber, highlights the transformative role of conversational AI in enhancing customer service interactions. She explains how Uber's customer obsession ticket assistant leverages natural language processing and deep learning to aid customer service representatives in efficiently handling support tickets. This AI-driven approach not only improves the customer care experience but also provides actionable insights for representatives, ensuring a seamless resolution process 1. Sam Charrington adds that the platform's flexibility supports various use cases, including research and prototyping phases, and integrates causal ML for uplift modeling and causal inference 2.

We have a lot of conversational AI data at Uber. One example is our customer obsession ticket assistant, which was one of our first use cases in this space.

--- Sam Charrington

   

Integration

The integration of conversational AI platforms with Uber's existing infrastructure, such as Michelangelo, is a strategic focus for Franziska Bell and her team. She discusses how newer platforms are built on Michelangelo's capabilities, while older platforms operate independently, highlighting the potential for future integration. This strategic alignment aims to streamline operations and enhance the scalability of AI solutions across Uber's diverse technological landscape 3. Sam Charrington notes that some platforms, like the experimentation platform, may remain standalone due to their unique methodologies and workflows.

We built on top of the Michelangelo capabilities and are actively utilizing this.

--- Sam Charrington