-
-
Jasmine(white shirt), Neha(purple jacket), and Vivian(gray shirt) developing a plan for the project
-
Jasmine hosting front end on her computer
-
Final project
Best wing award, Best domain name from GoDaddy registry, Best use of Gen AI, Best UI UX
Inspiration
We were inspired to create this chatbot because we wanted to create a helpful tool for new mothers. Motherhood is difficult as it it is both expensive and time consuming to see a doctor or seek advice from other people. We also wanted to address the lack of accessibility to healthcare for women. Moreover, there is a lot of information on the internet and it can get overwhelming to know which sources are reliable. Gaia makes it quick and easy to get reliable information at your fingertips.
What it does
Gaia is a generative AI chatbot, where you can type any question about parenthood, and it will generate a response. We built a chatbot using the google Gemini API key and we made our own programmable google search engine and linked websites from reputable sources such as Mayo Clinic, National Institute of Child Health and Human Development (NICHD), and more. From there users enter maternity related questions into the chatbot and it gives a response.
How we built it
We programmed the chatbot in python and used the Gemini API key. Google search engine ID, URL for search engine to feed in input, and then we used that URL to get responses from the search engine. We stored an API key in an .env file to make sure the API key doesn't get stolen. Moreover, we used Open AI to generate the initial draft for the front end and made changes in JavaScript and HTML. Then we used Flask to connect the front end and back end or handle the communication between the user and bot's response.
Challenges we ran into
Developing the front end and making changes was difficult because we were not familiar with the syntax for JavaScript and HTML. Initially, we wanted to use Typescript but found it hard to display locally, so we switched to JavaScript. It took some time for us to get familiar with Flask, as it facilitated the connection between the back end and front end. Furthermore, since Open Ai's API key required a payment, we had to find an API key that was free to use—Gemini.
Accomplishments that we're proud of
We are proud of developing a visually appealing front end with JavaScript and HTML and successfully using API keys to create a chatbot. Overall, we are proud that our functional project can help mothers and expectant mothers seeking advice.
What we learned
We learned how to develop a front end with JavaScript, HTML, and CSS. Also, we learned how to use API keys and pip install packages for this project.
What's next for GAIA (Generative AI Assistant)
We hope to fine tune it even more so it produces more accurate responses. Incorporating more reputable sources to generate information is also another goal in mind. For front end, we can organize the chatbot’s response better, creating a more pleasant user experience.
Log in or sign up for Devpost to join the conversation.