The inspiration from Cory came from the long waiting times on the phone to get information about Coronavirus testing where we live. Since information is also scattered across the websites of different departments and institutions, we thought it would be helpful to have a chatbot to take care of routine questions in an interactive and user-friendly way.

What it does

Cori is a universally accessible chatbot that serves as a one-stop-shop for all the basic information on Coronavirus. This would help save precious lives by increasing accessibility to information, reducing panic and misinformation and optimizing available medical resources. Ultimately, the goal is to relieve strain off the medical authorities by providing consistent, automatic answers to repetitive questions.

We believe Cori can help save precious lives by:

  • Improving accessibility to basic information regarding coronavirus

  • Relieve strain off from labs and medical authorities by helping people understand that not everyone needs to be tested for COVID-19. A lot of people panic and get tested despite not having little or not symptoms. Through a series of questions, Cori will help you understand whether you need to be tested or not? This can be extremely helpful in areas where there are a limited amount of "test kits" availabl (an issue faced in almost every region worldwide).

  • Helping combat the spread of fake news and bad medical advice through our verification cell

  • Helping disseminate the best precautionary measures and medical advice to the masses, collected from authorities like CDC (Centers for Disease Control and Prevention) WHO (World Health Organization)

  • Maintaining an index of local resources such as “where to get tested?” or a helpline etc. to direct you to as needed

How we built it

Cori uses the Google Cloud Platform (GCP) and Dialogflow to interpret questions and generate answers. Information in the current version is taken from the World Health Organization website, but this can be customized to suit local health authorities.

GCP and Dialogflow is a cost-effective and relatively simple way to develop and run a chatbot like Cory. This allows it to be a cost-effective solution that can be deployed in areas where budget constraints are a limiting factor.


As this was our first global online hackathon, we quickly realized that communication can be very difficult with team members spread across many timezones. Clear communication was challenging as we would sometimes have to work ahead without having the opportunity to receive answers from other members of the team. For any similar situations in the future, we will try to set internal "office hours" where everyone can agree to be available for at least a short amount of time.

A lot of cloud / technology providers have subscription fees and limited quotas for trial versions which prevented us from being able to perform an in-depth analysis of the available options.

Accomplishments that we're proud of

We’re a team of three complete strangers, spanning three different nationalities, who collaborated for the very first time on a common purpose of building something impactful through the power of technology. Although the development of the project is far from complete, we believe we’re onto something big and in due time we would be able to move our project into production.

What's next for Cori?

We are currently considering the following extensions to the Cori Project:

  • Integration with Facebook Messenger.
  • Mutli-language support (configurable via GCP).
  • Integration with Whatsapp (some governments are doing this already).
  • A “localized” index of doctors / labs / other resources where we can redirect users for further testing etc. Perhaps with Google Maps integration.
  • An easy integration SDK using which 3rd party entities can embed our chatbot into their website.
  • Integration with Google Assistant (configurable via GCP).
  • Integration with other platforms like a mobile application or twilio / helpline support (GCP takes care of speech-to-text conversion).

Built With

+ 20 more
Share this project: