Table 38
Inspiration
We can relate to the problem of waiting on the telephone line for an operator, all for only 3 minutes on how to solve a problem (either with the wifi router, a computer problem, etc). We saw Sigma's challenge and decided to develop a chatbot that could enhance the user experience for associates having trouble. Nevertheless, we also thought on making use of the data that the bot encounters and decided to create a companion dashboard with statistics regarding issues the associates are running into
What it does
It communicates with an associate through Telegram, solving common questions and issues that they may ran into in their day to day work. Botzilla also includes a dashboard to visualize statistics about the issues affecting the associates.
How we built it
We used Telegram's API in JavaScript, PHP for the database in MySQL, Python to communicate with IBM's Watson and HTML/JS for the dashboard.
Challenges we ran into
Getting to know the technologies, setting up the server, connecting the components and creating a good user experience.
Accomplishments that we're proud of
Everything is running in a server and in real time, the AI abilities of Botzilla and the potential for business impact.
What we learned
Work as a team, new frameworks, debugging in production.
What's next for Botzilla
Improve the AI and add more questions.
Log in or sign up for Devpost to join the conversation.