Name: ePIC HACK 2022 Teams Chatbot and notifier Team Falcon
Uploaded: 2022-06-03
Duration: 4 min 12 s
Description: ePIC HACK 2022 Teams Chatbot and notifier Team Falcon.

ePIC HACK 2022 Teams Chatbot and notifier Team Falcon. 

Scene 1

ePIC  HACK 2022 Teams Chatbot and notifier Team Falcon

Monitoring and Automating tasks are some of the most time-consuming operations that a user must perform for any product to be functioning at its best.. 

Scene 2

Monitoring and Automating tasks are some of the most time-consuming operations that a user must perform for any product to be functioning at its best.

Currently there is a multitude of dependencies and accesses that are involved in some processes – like the creation of GoCD pipelines – With the chatbot-based solution, we will be able to eliminate several manual steps and directly create the pipeline.

Diagram Description automatically generated. Value Proposition. 

Scene 3

Diagram

Description automatically generated

This implementation has far-reaching effects on the current procedures at Epsilon -. 

Scene 4

This implementation has far-reaching effects on the current procedures at Epsilon -

The current procedure to monitor regressions is a tedious one where a member of the team needs to constantly monitor the individual jobs and rerun in case of failures.

By using a teams webhook notifier and adaptive cards, the user can directly view/re-run/skip the execution of the failed stage/job of the pipeline.

If a user wants to navigate to a certain UI page, he will have to dig into his search history or fetch the link from a confluence page/notes page.

Instead, a chat bot can help him fetch the information and save valuable time.

If a user wants to create a GoCD pipeline, the current procedure involves a lot of navigation through UI screens and entering data.

The pipeline chatbot will be able to perform validations (which drastically reduces the probability of an error in the config repository) and create the pipeline without the user ever leaving the teams window!

This implementation can be adopted by anyone in Epsilon and they can create custom bots for their monitoring/automating/notification tasks.

The project was implemented using the Teams toolkit extension in VSCode and the Bot Framework Emulator to test the performance of the bot. The implementation was in python, but teams bots can be built in C#, Java and JavaScript as well.. 

Scene 5

The project was implemented using the Teams toolkit extension in VSCode and the Bot Framework Emulator to test the performance of the bot. The implementation was in python, but teams bots can be built in C#, Java and JavaScript as well.

The architecture overview of a simple bot is shown below. Once the bot receives a prompt it will fire an activity and the conversation will begin. Based on the logic implemented in the bot, the prompts will be fired to the user.

Graphical user interface, application

Description automatically generated

The pipeline chatbot was implemented using a waterfall dialog flow model as it takes in a fixed number of parameters as input to create the pipeline.. 

Scene 6

The pipeline chatbot was implemented using a waterfall dialog flow model as it takes in a fixed number of parameters as input to create the pipeline.

This solution can be extended to other tasks which require the user to enter answers to a certain number of standard questions.

In order to make the application globally available, it can be hosted on Azure services.

Graphical user interface

Description automatically generated with low confidence

Scene 7

It gets all the pipeline details from the user and automatically creates the pipeline. It confirms if the pipeline fails or passes successfully.. 

Scene 8

It gets all the pipeline details from the user and automatically creates the pipeline. It confirms if the pipeline fails or passes successfully.

Scene 9

The bot gives details on the tenant information, confluence pages the user is looking for, and walk through of product. 

Scene 10

The bot gives details on the tenant information, confluence pages the user is looking for, and walk through of product

Scene 11

Technological Challenges. We have hosted the chatbot using the bot framework emulator which is a local tool, in order to make it globally available, we will need to host the bot using an azure service.. 

Scene 12

We have hosted the chatbot using the bot framework emulator which is a local tool, in order to make it globally available, we will need to host the bot using an azure service.

We can extend this chatbot to an advanced model

where the chatbot adapts to the conversation and

generates its own answers or asks questions to

Meghana RK Mayank Rathod Avinash Sree Hari K Eshwar HS. 

Scene 13

Meghana RK Mayank Rathod Avinash Sree Hari K Eshwar HS

ePIC HACK 2022 Teams Chatbot and notifier Team Falcon

Scene 1 (0s)

Scene 2 (7s)

Scene 3 (28s)

Scene 4 (35s)

Scene 5 (1m 24s)

Scene 6 (1m 50s)

Scene 7 (2m 14s)

Scene 8 (2m 20s)

Scene 9 (2m 32s)

Scene 10 (2m 38s)

Scene 11 (2m 50s)

Scene 12 (2m 56s)

Scene 13 (3m 17s)

Scene 14 (3m 26s)