Ate

An older sister for the eldest daughter (note: it's for everyone)

Comment

Inspiration

The challenges for this hackathon were all inspiring. The one I focused on is the empowerment of women.

Sometimes, all we need is an older sister to rant to. "Ate" (ah-teh) is the Tagalog word for "older sister".

The inspiration for this chatbot stems from the unique role of eldest daughters in many families. They often find themselves in a position of both responsibility and emotional support, often feeling the pressure to be strong and perfect.

When the eldest daughter is stressed, they need an older sister to talk to as well. This chatbot aims to provide a safe, non-judgmental space for these individuals to express their emotions without the fear of burdening or worrying others.

Though the inspiration for this project are eldest daughters, the "Ate" chatbot is the older sister to everyone.

What it does

The chatbot takes on the personality of an older sister while chatting with you. Communicating with it will be like talking to a friend in a social messaging platform. Its responses will be short and broken down into chat bubbles to become easily readable and less intimidating compared to a giant block of text. The site's UI colors are similar to the popular messaging apps too.

You may talk to it about your day or just talk about your feelings. If you feel bad, you can also ask it for remedies.

How we built it

Website The chatbot website is a Nuxt JS project. Tailwind is also used.

Gemini Gemini was used during the brainstorming and research part of this project. It gave the initial system instruction prompt for creating the older sister persona. More instructions were added to the prompt as development continued. Gemini also provided the utility functions needed in the code.

Gemini API Gemini API is the main component of the chatbot. After following the "Gemini API quickstart" guide for NodeJS, I followed the Build an Interactive Chat example in the docs.

Assets Icons are from SVGRepo under a public domain license.

Challenges I ran into

I could not use the Grounding with Google Search feature in my chatbot because of a rate limit error.

Accomplishments that I'm proud of

I am proud of learning how to use the Gemini API to build a chatbot. I am also proud of myself for participating in this hackathon.

What I learned

  1. I learned how to use the Gemini API and how powerful it is. I also learned how easy it is to create a chatbot using it.
  2. I learned about the different AI tools of Google
  3. I leaned about the UN SDG 5: Gender equality and it made me aware of the work that needs to be done.

What's next for Ate Chatbot

  1. Allow the user to set the chatbot's name to make it feel more personal.
  2. Allow the user to set their "location" which can be used by the bot as reference for their answers. For example, if you live in Country A, you would want product recommendations from Country A instead of Country B.

Built With

Share this project:

Updates