PullString brings more power to developers

July 30, 2017, 3:55 p.m. By: Prakarsh Saxena


PullString was founded in 2011 with the vision of revolutionizing computer chat bots conversation through text and voice by incorporating arts and science along with intuitive design and technology. The company’s primary product, the PullString Platform, offers a professional authoring environment, hosted AI and Machine Learning Runtime, and integrated conversation analytics to various verticals like gaming and VR, travel, automotive, health and fitness, finance, human resources and IoT. With direct support for all major conversational platforms like Messenger, Amazon Alexa, Skype, Slack, Kik, mobile and the web alike, the PullString Platform has been used to create award winning experiences like Call of Duty’s Lt. Reyes on Messenger (pretty cool!), Hello Barbie and The Winston Show.

Rule Based AI and ML- Recipe for a smart bot

PullString enables the developers and content creators to create intuitive and conversational bots as fast as possible. The most compelling of all is the fact that the interactions are mostly self- learning based. It uses a combination of rule based AI and ML to create conversations which are appealing to the general audience and seem genuine and useful, also maintaining the fact that the conversations need to be appropriate with regard to the context. This can make intelligent conversations, about precise or off- topic things, a reality. The Hello Barbie product by PullString can let kids talk about anything to the bot. Similarly, the Call of Duty bot can also talk in the same way how the gamers communicate. They did this by pointing the AI at a large CoD subreddit and the bot learnt the language used there to enable the bot speak the same way as the gamers would in the same tone and style.

PullString is growing tremendously in the race to become the pioneer in the field.

  • Provides a professional authoring environment, where the developer can define their own context- specific intents or use the pre- built ones in the library, along with integrating text, images, audio and video, structured messages, cards etc.

  • A highly intuitive conversational AI platform using a mix of ML and Rule Based AI, using intents to enable modeling the high- level goals or desires of the users, with proper synonyms in the dictionary and dialog management.

  • An integrated Chat Debugger in the PullString Author to test conversation at any point

  • Extending capabilities of the bot like accessing news, weather, knowledge base etc. using the Web Services.

PullString brings more power to bot developers through its various methods:

Web API: Using the PullString Web API to add conversation to your own applications, websites or devices and the API takes care of the NLP, AI and ML for the code.

SDKs: Leveraging the open source SDKs for iOS, Android, Python, JavaScript and Unity.

Code Execution: Dropping Node.js code directly into the bot’s content to get complete control, include full access to the conversation state.

Integrations: Easily integrating existing intents already built in other services like API.ai, LUIS, or Wit.ai

With such wide application of the platform, and its efficient and optimized text, voice and VR capabilities, PullString will sure tap its way into the huge potential market of IoT and voice- controlled devices into the future.

PullString Platform Introduction

Video Source: PullString, Inc