Chatbot for DevOps Department

The chatbot was created by our DevOps Department to solve several communication/management problems and to automate repetitive actions, increasing the speed of releases.

It all started with curiosity. We aimed to improve internal communication, making it faster and more effective; that is why we created our own enterprise chatbot. This chatbot was developed in Python and had several libraries integrated into it. The team placed a bet on the Telegram messenger because it’s one of the most popular and secure, but in reality, any messenger, such as Viber, Slack, or Skype, can be used.  

Some of the Goals We Achieved with the Help of This Chatbot: Collection of Performance Statistics

Our chatbot is able to collect performance statistics. For example, if the last ten tasks failed or were successful, the statistics are displayed in red or green, respectively. At the end of the month, a detailed report is automatically created as a file for all projects and their participants.

Problematic Situations Prediction

The possibility to predict problems is one of the most valuable features of this chatbot. This feature allows fixing the problem in advance. We can also monitor the activity of users of our products and analyze this data. This gives us the opportunity to choose the most suitable time for new releases.

Remote Task Management

For instance, there was a need to copy the database, but you don’t have access to a PC at the moment. Using a chatbot, it’s easy to send this command via messenger and get things done remotely.

Security

Of course, we didn’t neglect security. As the messenger has limited access, only those who passed the authorization can join and send commands, so the security part is well-thought as well. 

Timely Reporting

After receiving this task, the bot is assigned this simple task and is ready to perform it and report about it. If the task was not successful, it explains the reason for the failure. If it was a success, the bot informs about the details of the task. For instance, it can send the directions to the folder where it is preserved in the database. 

Mundane Tasks the Chatbot Can Perform:

• New releases creation;
• Release deployment;
• Test starts;
• Backup of the database;
• Monitoring the phase of development and notifying the team about possible future problems. For instance, if there is not enough memory on some storage, the bot can warn the team about it. 

Conclusion

So internal chatbot is the ultimate solution for the operational problems common to all teams. We had this engineering problem, and we’ve solved it using an engineering solution. So far the effectiveness of our performance has increased by 17% as we automated the tasks that were redundant. 

Languages Used: Python.