Gaia – Goddess-Guided Mentorship

Gaia Guided Mentorship

Inspiration

Our team was inspired by our own experiences of being women in tech at NJIT. We created this platform to make sure no woman feels alone. Our goal is to build a supportive community where women can find mentorship and access the resources they need to grow with confidence. By matching users with a goddess based on their needs, we hope to remind every woman that she carries the same divine power within herself! 🌿

What it does

Gaia is a web application with a dynamic AI chatbot designed to support women on campus by connecting them with resources, guidance, and community! 🌟 Inspired by Greek goddesses, the platform begins with a chat with Gaia—the source of all life. From there, users can connect with the goddess they need most: Athena for wisdom and academics, Artemis for mentorship and independence, Aphrodite for confidence and empowerment, Tyche for scholarships, or Gaia for general support.

Each goddess searches the web to provide relevant answers and resources—whether it’s events from Highlander Hub, research being conducted at NJIT, tutorials from YouTube, tutoring services on campus, and more. Students can log in securely, access curated NJIT resources, and find encouragement through a goddess-inspired mentorship experience that ensures they’re never alone.

How we built it

Frontend: React, Vite, HTML, CSS, JavaScript, TypeScript

Backend: Python, Gemini AI, Azure AI, MongoDB, Auth0, FastAPI, BeautifulSoup

Challenges we ran into

A lot of the softwares used in this project was new to most of our team members, including Azure AI, Google Gemini API, Mongo DB. This hackathon's challenges challenged us to go outside our comfort zone and take on new softwares, and there definitely was a learning curve in the beginning. With the help of workshops, company representative, the WiCS team and AI + research, we were able to integrate these softwares into our project. Going into the technicalities, we had trouble scraping + tagging the content from the website: which was crucial for our goddess-switching feature to work. Additionally, in the front end development we had problem with alignment of the visual components, and general user interaction with the backend protocol/ chatbot.

Accomplishments that we're proud of

We're really proud that we took on a project that pushed us all to grow and step outside our comfort zone, and it was really satisfying seeing the backend and frontend come together. We’re proud that our work aligned with a mission of empowering women in male-dominated fields.

What we learned

We learned a lot about implementing the new softwares, more specifically Google Gemini API (chatbot, intent classification), Azure AI (RAG implementation), and MongoDB (chat history). None of us had worked with Gemini API before so we had to learn how to implement it and prompt it within our code. With Azure AI's Cognizant Search, we turned scraped data into vectors that integrating with our chatbot to give specific and contextually relevant responses. MongoDB enabled us to manage user sessions and store chat history.

What's next

We will introduce personalized pathways to help students with career roadmaps, wellness check-ins, and academic planning. The platform can grow beyond NJIT by tailoring resources and guidance to local events at other campuses. To keep students engaged, we’ll add simple gamification features like badges and streaks that encourage regular use and long-term commitment. We will also expand the web app to include more details like an about us, notes and planning section, NJIT schedule planner integrated into the web application etc.

Built With

Submitted to

GirlHacks 2025
- Winner First Place

Created by

I worked on python scripts to scrape and process the data so it could be used by our RAG, built with Azure AI. I incorporated Gemini API to tag the data more intelligently. I also contributed to the design process and helped plan and manage the project.

Nandika Karnik
I worked on the front-end. I used React, HTML, CSS, and JavaScript, which I was familiar with before. I also used Vite, which was something new for me! I designed how the website would look, the color scheme, the art style, and more.

Marcela Moura
I worked on integration between the backend and front end. It was incredibly difficult to manage the complex interactions needed to switch between goddesses, but ultimately very rewarding to see it all work.

Assaf Izhar
Computer Science Student at NJIT
I worked on scraping data from NJIT websites and processing the data to be used by our RAG, built with Azure AI. I also worked on connecting the scraped content to Azure AI (indexing). I helped with Gemini API to tag the data. I also contributed to the design process.

Nidhi Sakpal