Casia: Smart Plant Care Assistant

Your digital botanist providing tailored plant care advice, reminders & answers to your queries, all in one easy-to-use app.

Comment 1

Inspiration

In the midst of urban living, I found solace in cultivating my own green space. However, the journey to successful urban gardening was fraught with challenges. From finding reliable plant care information to remembering various plant care tasks, the road was often bumpy. Recognizing these everyday struggles among urban gardeners, the idea for a digital solution - Casia, the Smart Plant Care Assistant - was born.

What it does

Casia is more than just a mobile application; it's a personal botanic advisor readily accessible on your phone. By integrating artificial intelligence models and cloud resources, Casia provides tailored plant care advice, tracks plant care tasks and milestones, offers nurturing tips, and shares knowledge to foster your growth as a plant parent.

How we built it

We leveraged our expertise in data science and machine learning to integrate AI models and cloud resources into a user-friendly phone application. We created a database of plant care information using Pinecone's vector database and designed our language model to understand and respond to user queries effectively. At the core of it, we have a cognitive architecture with hierarchical memory that refines knowledge and builds more context which enhances its ability to provide more specific relevant answers.

Challenges we ran into

The process of integrating AI and ensuring the model correctly interpreted and responded to a wide range of user queries was challenging. Latency for AI models has been a painful thing to work with and some of the best models are the slowest which requires us to compromise. Moreover, creating a user-friendly interface that could seamlessly deliver a wealth of information was a task that required careful planning and execution. My team is small and we are busy with work and family so this really has been a hard-working week. Our dark eye bags can attest to this.

Accomplishments that we're proud of

Successfully combining artificial intelligence with practical, user-friendly application design is something we are very proud of. We've created an application that provides invaluable help to urban gardeners and promotes sustainable living and mindfulness. I think my team has their first experience working with LLMs and vector databases which is something they enjoyed.

What we learned

This project taught us the importance of user-centric design and the potential of AI to simplify complex tasks. We also learned the importance of careful data management and the immense value that can be derived from accurately interpreted data. So much data doesn't mean a better user experience, but surely meticulously curated data is.

What's Next for Smart Plant Care Assistant

We envision Casia as an integral part of the urban gardener's toolkit. We plan to continuously update our plant care database, refine our AI model to better answer user queries, and incorporate additional features such as a community forum where users can share their experiences and learn from each other. Bringing better recommendations of content and plant care tips based on user preference and environment. We believe Casia has the potential to revolutionize urban gardening, making it more accessible and enjoyable for everyone.

Built With

Share this project:

Updates

posted an update —

Pinecone doesn't work with all the values that are generated by some transformer models. This will take time to work through as well, so I ditched huggingface transformer models and used openai as they work out of the box and can insert to pinecone without issues. (they incurr cost but speed is important)

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