Now What?!

Inspiration

With so many burning deadlines for university students, it's getting harder to do what you like when you feel like it. Unlike regular task schedulers. If only someone could always understand us and give truly personalised ideas! Introducing Now What.

What it does

Now What cares about your mood better than your closest friend. It searches for data from Moodle automatically and gives YOU suggestions based on your mood. YOU are always in control of your life, and Now What learns your mood and balances deadlines with mental health.

How we built it

Multi-modal LLMs were used to evaluate the user's mood and task difficulty. The user would then be able to set the weight of each factor. Recommendations are refreshed every time the user works for two hours.

Challenges we ran into

Even though the backend codes were solidified over the course of a day and a half, we suffered from front-end GUI issues as none of us had any prior experience. LLMs also don't always give the same result format, and we have to accommodate. We also need to wait for the LLMs to reply with the response.

Accomplishments that we're proud of

We tamed LLMs, used the multi-modal capability of ChatGPT, and developed a voice and text assistant that is highly personalised for each user. Knows the personal context and maximises the mental health.

What we learned

Interacting with AI. Combining hard code with AI codes and prompts.

What's next for Now What?!

Develop a fantastic and caring GUI that maximises user experience. Improve mental health maximisation algorithm. Connect to real APIs of actual Moodle pages (for example, the vle of University of Canbridge)