Inspiration:
As a college student, there are many things that are expected from us and that we expect from ourselves. We have a lot of tasks to get done that we don't even know where to start. Even though people pick tasks in a random manner and try to see if they can finish all their tasks, the path they take to finish them would be a long and strenuous path. Meanwhile, if they had their tasks sorted out, they would be able to properly allocate time to the different tasks, which would help them complete their tasks at a quicker pace and, at the same time, help them improve productivity in all they do.
What it does:
This AI agent takes in a set of tasks as a paragraph or a list, as long as the different tasks are separated by a period. It then analyzes the tasks one by one. It analyzes what each task entails and checks them against each other, then arranges them from the most important to the least important. it factors everything stated in each task and keeps them in mind whilst;t it rearranging the tasks. It also arranges tasks in real time. If after I put in a set of tasks, I come back later remembering some tasks I omitted and I add just those new tasks, it would include those new tasks with the old ones and rearrange the entire list into a new list of sorted out tasks.
How we built it:
I built it using the Agent template on Vertex AI builder. I chose this template so that it would be accessible to the public and people could test it and run different scenario tests on it. I gave the agent a well-thought-out and well-defined set of instructions to follow when carrying out the tasks. This allows it to function appropriately and properly.
Challenges we ran into:
One of the significant challenges I ran into was understanding the layout of the platforms. It took me a while to get the hang of it as I had never used either platform before in my entire life. I also had an issue with the AI response and result generation. It would give the wrong results and two different results. Lastly, I had an issue with making the AI accessible to people. I could not integrate it with platforms like Google Calendar or Microsoft to-do list so I had to make it a website so that people would be able to access it and try it out.
Accomplishments that we're proud of:
I am proud that I was able to accomplish something I never thought I'd be able to. I could have turned a blind eye to this hackathon and walked away from it, but the thought of learning something new in technology drove me on. And thanks to that drive, I have opened up a new pathway for myself that I can dive into whenever I want.
What we learned:
I learned a lot about the Google Cloud Platform and the Vertex AI builder. I also learned about cloud functions, and my understanding of APIs became deeper.
What's next for Task Organizer:
My original plan for the task organizer was not for it to be an AI where we input our tasks and it helps us sort them. I planned to integrate it with systems like Google Calendar and Microsoft, where it would receive the tasks from them, sort out the tasks for us, and return the result as a new set of tasks on the interfaces. After this, I plan to continue on that path and see how far I can integrate this system into platforms such as those and how much I can learn from the AI world, especially for specialized AI systems.
Built With
- google-cloud
- python
- vertex
Log in or sign up for Devpost to join the conversation.