kAIron is a web based microservices driven suite that helps train contextual AI assistants at scale. It is designed to make the lives of those who work with AI-assistants easy by giving them a no-coding web interface to adapt, train, test and maintain such assistants.
kAIron is currently built on the RASA framework. While RASA focuses on the technology of chatbots itself, kAIron, on the other hand, focuses on technology that deals with the pre-processing of data that are needed by this framework. These include question augmentation and generation of knowledge graphs that can be used to automatically generate intents, questions and responses.
It also deals with the post-processing and maintenance of these bots and provides actionable metrics regarding bot performance.
(kAIron’s still in the beta phase. We’re actively developing it – click here if you’d like to try it out.)
One of the biggest barriers to the adoption of chatbots within organizations is adapting contextual AI assistants to specific domains. This means extensive work creating intents by going through documentation, testing accuracy of responses, etc. kAIron’s aim is to provide a no-coding self-service framework that helps users achieve this. These are the features in the 0.1 version –
kAIron is built for two personas:
kAIron’s released under the Apache 2.0 license. You can find the source here – https://github.com/digiteinfotech/kairon. Our team’s current focus within NLP is Knowledge Graphs – Do let us know if you are interested.
kAIron gives you the ability to quickly train your chatbots by allowing you to add intents , add training examples and their responses.
After training you can directly download the bot files or you can deploy to a website immediately. kAIron gives you the ability to train your chatbot and then test your chatbots in-situ. This reduces the turn around time from chatbot creation to testing drastically.
kAIron also features a dedicated metrics view to show you how your chatbot performed historically.
kAIron’s still in the beta phase. We’re actively developing it!