Productivity Manager SlackBot

Inspiration

We wanted to try something that had AI and machine learning involved, as it is a topic that we are all interested in. We saw the challenge from Business-In-A-Box, and thought it would be perfect for us to learn a lot, and try to build something really cool.

What it does

Our chatbot has many functionalities. It is automated to be capable of sending out custom messaging at a scheduled time, to few or many users. To manage productivity it begins each day by sending out a status report of each person in the slack, and at the end of the day it sends a request for a progress report which it then stores in a database. Our chatbot can also update the project of an employee in the database, it can get the status report of any employee. Our bot is highly interactive with the slack workspace owner, and a great motivator for all other users!

How we built it

We built our chatbot using Python, MongoDB and JSON. We also used a Slack API and Flask for the chatbot itself.

Challenges we ran into

It was difficult getting started. We had a lot of goals of all the features we hoped our bot would have, but after a short awhile it turned out a lot of those seemed too ambitious. It was also very difficult connecting Python to MongoDB, and connecting the SlackAPI too Python. Those parts of the project were much longer than expected.

Accomplishments that we're proud of

We are very proud of the functionality our bot ended up having. It can perform many tasks, and even interact with a user. We weren't sure how far we would get, and in the end we got a lot more accomplished than we thought we might have. We all had separate tasks, and we were working on different functions, and we were worried it would be difficult to bring it all together. But, we managed to be able to parse the input from the user from Slack, into input for a Python function, that would then go to MongoDB. And this was something we were worried we wouldn't be able to do, so we are proud we got the workflow across the different applications to run well.

What we learned

We learned a ton about MongoDB, as well about using the SlackAPI. And how to connect Python too many different sources, and be able to run processes through different applications. For instance, being able to get a message from Slack, parse that message into input for a Python function, and then run that function to the Mongo database. We all feel much more skilled with Python, as well as using MongoDB and the Slack API.

What's next for Productivity Manager SlackBot

It would have been nice to get more user data, and be able to implement some machine learning, where our bot can learn what employees are more productive than others. This way, it could maximize which employees are assigned to which tasks, and would benefit productivity in the workspace.