💡 Inspiration
The inspiration behind Task Craft is rooted in the need to simplify and streamline the process of transforming thoughts into actionable tasks. Students today are already bothered enough by a pile of assignments. With Task Craft, you can create a personalized calendar schedule that is both time-specific and has self-priority functions.
🚀 What it does
Task Craft is an AI-powered tool designed to turn your thoughts into Google Calendar tasks effortlessly. It addresses the common issues faced by students, such as the overwhelming nature of post-class responsibilities and the difficulty in accurately estimating the time required for assignments. The tool allows users to dump their thoughts, and the AI, powered by Google Palm 2 AI and Vertex-AI, creates Google Calendar events based on the provided information. If you want to do the simpler tasks first, simply tell the AI that you want to do the simpler task before others, and it will create a calendar schedule that suits you best!
⚙️ How we built it
Task Craft is built using a cutting-edge stack of technologies. Python and Streamlit provide the backend and user interface, respectively. Google Palm 2 AI, through Vertex-AI, enhances the natural language processing capabilities, while Whisper ASR handles real-time audio transcription. Lang Chain supports multilingual functionalities, and the Google Calendar API seamlessly integrates the generated tasks into the user's calendar. Custom CSS is employed for a personalized and visually appealing user interface. The entire system is hosted on the Google Cloud Platform for scalability and reliability.
⚠️ Challenges we ran into
Building Task Craft presented several challenges, including integrating multiple APIs, achieving real-time audio transcription, and designing a user-friendly interface that provides clear feedback. One of the significant challenges was ensuring the correct output from the Google Palm 2 AI every single time. We implemented a double-layered approach, using Vertex AI initially to generate the output and then employing another Vertex AI instance to verify if the output is in the correct format. This approach ensured accurate results for a wide range of user inputs. ANOTHER problem: Most of us are first time hackathoners, trying out Streamlit for the first time. Having a background in basic python helped but it was really tough getting around the documentation for the components.
🏆 Accomplishments that we're proud of
Despite the challenges, we're proud to have successfully integrated the latest technologies, created a real-time interaction experience, and prioritized a user-centric design. Task Craft provides users with a cohesive and functional solution for managing their schedules effectively.
📖 What we learned
The development of Task Craft provided valuable insights into API integration, natural language processing, and user experience design. Working with Google Palm 2 AI, Vertex-AI, Whisper ASR, and other technologies on the Google Cloud Platform deepened our understanding of these advanced tools and their applications in real-world scenarios.
🧐 What's next for Task Craft
Task Craft will continue evolving to address user feedback and incorporate advancements in natural language processing, AI, and scheduling technologies. We are committed to staying at the forefront of technology to provide an even more seamless and intelligent scheduling experience. Feel free to adjust the details based on your project's specifics and any additional information you'd like to include!
Built With
- custom-css
- google-calender-api
- google-cloud
- google-palm-2
- lang-chain
- ngrok-server
- python
- streamlit
- vertex-ai
- whisper-asr

Log in or sign up for Devpost to join the conversation.