Wit.ai at F8


Thanks to everyone who stopped by our booth! It’s was nice to meet you and see all the great products, platforms and companies you’ve been building on top of our API! Thanks for the great feedback and the contagious enthusiasm.
The Wit.ai team also hosted a breakout session and a classroom talk!

Breakout Session — Using Natural Language Processing to Enhance Conversations

Led by Software Engineers Anthony Kesich and Julien Odent, this session covered how you can use NLP to enrich conversations between people and your bot. You can watch the full recording here.

Highlights included:

  • An overview of M Suggestions and how the virtual assistant helps enrich people’s conversations and get things done.
  • An overview of the technology behind M Suggestions and Wit.ai.
  • How developers can easily leverage NLP with Wit.ai to reduce user error, scale and automate tasks.
  • A walkthrough of the newly launched built-in NLP integration with the Page inbox to enable Page admins to easily create a custom Wit app and then automate responses to common requests using anonymized samples from past messages their Page received.

Classroom Talk — Duckling: An Overview of the Open Source Probabilistic Parser

Duckling is a Haskell library that parses text into structured data. In this talk, Julien Odent deep dived on the technology behind Duckling and how to add in your own language.