Chatbot to Live Operator Switch.

About project


Consider system, which includes bots and live operators (e.g. a complicated support infrastructure). The problem which arises in such a system is to determine the appropriate moment of switching user from to live operator.

Solution description:

Application which consists of four components: chatbot (e.g. Telegram), backend (Chatbot/Live Switchover, ML model), module which is responsible for chatbot conversation logic (e.g. via integration with DialogFlow) and module which provides connection to live operator (e.g. Intercom).

It works as following:

  1. User is some dialog with bot, where dialog logic was derived from DialogFlow.
  2. Chatbot/Live Switchover ML model is a background task, which is listening to and analyzing conversation to detect the moment when user is needed to be switched from conversation with bot to live operator.
  3. At the moment ML model triggers switcher, user is automatically connected to live operator.



Yes, we are building own projects!


"Doing business is not a trivial task. You can create a good product but get zero customers. That's because development is the important part of building products, but marketing and sales are essential. If you a software development company you aware of technologies and architecture, not customer business needs. That's why a lot of collaborations between businesses and software development companies failed.

In DevRain we are building own products. It helps us to go through all the process - from the idea to the market like our clients do. We learn a lot how to fail, how to find first 1000 customers, how to be paid for them, how to work with media and opinion leaders. That's why we are building high-quality products for our clients because we understand them and their needs.

Let's make great products and business together!"

– Oleksandr Krakovetskyi
CEO, Co-founder

All customer projects →