TaskTrekker

• 

  sample conversation

• 

  sample data

Inspiration

The concept behind "TaskTrekker" is inspired by the popular anime "Solo Leveling," where the main character is assigned daily tasks by a mysterious system, helping him grow stronger and unlock new abilities. In the anime, the protagonist's journey of constant improvement and overcoming challenges captivated audiences, demonstrating the power of daily, structured tasks in achieving significant progress. Similarly, TaskTrekker aims to be your personal growth companion, assigning you daily tasks tailored to your interests and skill levels. By integrating this engaging and motivating approach, we hope to help users experience the thrill of continuous improvement and the satisfaction of achieving their goals.

What it does

TaskTrekker is a self-improvement app that helps users achieve their personal and professional goals through customized daily tasks. Users select their areas of interest, such as gym workouts and data science, and choose their preferred difficulty levels. Based on these preferences and their progress, TaskTrekker generates daily tasks that are both achievable and challenging, ensuring steady growth and sustained motivation. The app also tracks user performance over time, providing feedback and insights to help users stay on track.

How we built it

TaskTrekker was built using Vertex AI Agent, which played a crucial role in the development process. We started by feeding Vertex AI Agent with unstructured data consisting of sample everyday task conversations between users and the agent. This dataset included interactions where users expressed their preferences, selected difficulty levels, and provided feedback on completed tasks.

By processing this rich dataset, Vertex AI Agent was able to understand the context and nuances of user interactions, learning to generate personalized daily tasks that align with individual preferences and performance history. This approach allowed us to create a robust system capable of dynamically adapting to each user's unique needs

Challenges we ran into

Building TaskTrekker presented several challenges, including preparing and integrating unstructured user-agent conversation data . iteratively improving the agent to get desired results took lot of work.

Accomplishments that we're proud of

We are proud of several key accomplishments in the development of TaskTrekker. First, we successfully created a Vertex AI Agent that adapts to individual user needs. We also developed a clean and intuitive user interface that enhances the user experience. Moreover, we integrated comprehensive progress tracking features that provide valuable insights and feedback to users, helping them stay motivated and on track with their goals.

What we learned

Throughout the development of TaskTrekker, we learned the importance of user feedback and iterative design. By continuously testing and refining our features based on user input, we were able to create a more effective and enjoyable agent. We also gained valuable insights into google agent builder.

What's next for TaskTrekker

We plan to further enhance TaskTrekker by adding evaluation metric functionality. This will involve asking users specific questions related to their tasks or verifying outputs such as code submissions to ensure task completion and understanding. Additionally, we aim to introduce a points system where users are rewarded with points for completing tasks and may receive penalties for not completing them. This gamification element will help increase motivation and engagement, providing users with tangible rewards and a sense of accomplishment as they progress. These enhancements will make TaskTrekker even more effective in helping users achieve their personal and professional goal

Built With

• google-cloud
• vertexai-agent