Yes, the Kore.ai Bots Platform includes our Bot Builder, an intuitive web-based tool with a repeatable process to design, develop, test, deploy and continue evolving chatbots at enterprise scale, without code, custom software, significant server space or major changes to your infrastructure.
The Bot Builder tool is easy to use for anyone with web-services knowledge, and typically requires very requires little ramp-up time to get familiar with the features and navigation of the system. The interface is highly intuitive and designed to be used with minimal guidance and training. The primary elements involved in building bots are explained at a high-level below:
- Designing the bot and the tasks it will perform – Building a chatbot begins by defining the bot and specifying what the bot will do for its users. The Bot Builder offers developers flexibility in defining the types of tasks their bot will perform: take action, deliver alerts, answer questions, produce on-demand or scheduled reports, or execute workflows. This includes being able to turn complex interactions and decision trees into conversations with the Dialog Builder element of the tool.
- Configuring and training the bot for NLP – The Bot Builder includes a step for configuring and training the bot to engage users in natural language. You select the chosen intent identification mode as either a hybrid model of Fundamental Meaning and ML, or ML-only. The Kore.ai Platform “auto-NLP” enables the bot used on the task name and the naming conventions that Kore.ai recommends, then configure and train the NLP with additional defined synonyms and patterns (or idiomatic ways or saying things or slang) for Fundamental Meaning, and/or with known user utterances for ML. Standard responses for questions, statements, errors, and warnings can also be randomized and varied by channel to provide a rich, personalized experience.
- Adjusting settings – Developers can adjust settings for the bot, including areas like authentication and channel deployment. Stored authentication profiles and full access to header data make bots quicker to deploy on Kore.ai’s Platform because developers can easily reuse existing profiles for future bots. Omni-channel deployment is also easy and fast. You design the underlying bot tasks once, and then just select the channels for deployment in the tool. However, the developer can still differentiate cross-channel experiences by altering message responses or leveraging channel-specific UI elements like date selectors, carousels, and more – without having to re-write all the underlying components for each channel.
- Testing throughout the development process – A developer can validate and test authentication settings, API calls and data capture, response message generation, and NLP intent and entity recognition as they build for quicker resolution of issues. Test request chaining, intent recognition, conversation flow, and use the Kore.ai Dialog Builder debugger to expose chat logs and outputs in real-time as a conversation with a chatbot is happening. Once the building blocks are tested and verified, the developer can also test the complete end-to-end behavior by invoking the bot and conversing with it. Useful trace and log messages are also provided in each of these steps to help the developer debug potential issues.
- Deploying bots – Once a bot is ready for primetime or after an associated task has been upgraded, developers can publish that bot or associated bot task to the enterprise Bot Store or directly to end-users. Enterprise administrators are sent a notification to review and approve a bot before it can go live, ensuring that all bots maintain company standards.
Kore.ai can help customers build bots, and we offer extensive product documentation on our developer website at https://developer.kore.ai, which provides context-specific help from within the Bot Builder tool.