SFL Botathon 2017: Groom-bot and the Idea Behind it
As the guys enter the room, one can say they are pretty excited- there’s new technology introduced by them that pretty much happened overnight and was voted as the best project during the Hackathon event.
A little Background
SFLians conducted a Botathon on November 17–18 where 5 teams introduced innovative bots and competed for the winning place.
The five competing teams were:
- Secret Bot
- Attendance Bot
After working on their bot ideas for 24 hours, the teams took turns to introduce their bots using presentations and the winner, Groombot team, was picked.
When asked to tell about the idea, how it was conceived and initiated, Vardges and Khachatur had quite a bit to say:
“Creating a bot that would alleviate grooming was an idea that I had for a long time, ever since I joined SFL. As I was participating in weekly plannings, I noticed that our Product owner could give approximate estimates to tasks without having a technical background. This is something that led me to believe that using AI could significantly simplify this process relying on previous estimations and language analysis”.
When I joined SFL, Vardges soon asked me to team up with him in refining this idea, as we had previous successful experience working together. To make this a learning experience, I picked technologies such as ReactJS and NodeJS that I had no previous experience working with.
I’ve long had interest in Machine Learning and AI, so working on this project was the first chance to put the knowledge I had accumulated into practice.
Running a Botathon is really a cool idea as it helps enhance team spirit and is a strong team building tool.
At the same time, it shakes up the routine office life and combines elements of competition and fun! Victories like this make you think that you are on the right way to reaching your goals!
During Botathon, Vardges and Khachatur presented the MVP for Groombot.
This tool is able to make initial predictions for tasks based on previous estimations, arrange task division, categorization, complication analysis and more. It is not only useful for POs for their day-to-day tasks, but also for people without any technical background- like the CEO trying to estimate how long it may take to deliver a task or a project.
The Future of Groombot
Eventually, it comes to the plans for the future. On that end, the guys are very positive- Groombot has lots of potential that can be unveiled. As Atlassian plans to integrate machine learning tools into Jira, Vardges and Khachatur imply that Groombot can become a successful add-on to the platform.
Besides, they are planning on adding webcam functionality that can calculate the time spent working every day, and eliminate the need of manually entering these values every day.
As we finish the discussion and return to the go-get-things-done workflow, let’s look at some of the moments we captured during SFL’s Botathon!