Inspiration

In life, we're always faced with many decisions. Whether it's at your work, school, or just simple decisions like what to eat, we've always had to make choices regarding what we want to do. Some decisions are easy, but there are also the harder decisions, the ones that can alter major aspects of your life. When it came time for me to make those decisions, I've always wished I had someone that I can talk to, that had as much clarity about my life as I do. That's where Decido comes in.

What it does

Decido is an application that allows users to input their major life decisions and log them. Users will be able to come back and revisit the application to see what they did for these decisions and the thought process that went behind it. The Users will also be able to chat to an AI agent, which then gives advice to the user on future decisions, based on past decisions that were inputted into the application.

How we built it

I utilized Gemini 3 and the Google AI Studio to completely built this application from the ground up. I specified three features. Firstly, I want Gemini intelligence in my application to allow for the agent to perform a multitude of tasks when handling user inputs. Secondly, I also want an AI-powerd chatbot that would be context-aware and would remember conversations, which would enable the agent to provide advice to users based on the input. Lastly, I also want the agent to Think more when needed, which would push the agent to think if it deems the situation as complex.

Challenges we ran into

There were a lot of challenges that I ran into. The main issue, however, is crafting the entire mental model for this application from the scratch. As this was a novel idea, there were no references that I could use to go off. I had to create a lot of processes from scratch, but I also had to make sure that those processes work for the betterment of the user and will provided the value proposition that Decido intents to provided.

What's next for Decido

The next steps would be to improve and finetune the processes to help get better insights out of the AI agent. I would also like to create a voice version, where the user can just talk to the AI and the agent would still be able to provide the same benefits as the chat version.

Built With

Share this project:

Updates