Plantbook

Inspiration We were inspired by the idea of bridging the gap between plant enthusiasts and professionals. While AI models can answer many questions, sometimes they fall short — especially with nuanced or local plant care advice. Plantbook aims to fill that gap by connecting users with verified professionals when the model can’t provide a confident answer.

What it does Plantbook is an AI-powered plant knowledge platform. Users can ask any plant-related question — from identification to care tips — and our AI model tries to provide an answer instantly. If the model’s confidence is low or a question needs expert input, the query is seamlessly escalated to a community of plant professionals who provide personalized answers.

How we built it We built Plantbook using a combination of modern web technologies and AI. The frontend is built with Next.js and Tailwind CSS for a clean, responsive UI. Our backend integrates a fine-tuned language model for initial question answering. When the AI cannot answer with high certainty, it triggers a fallback that routes the question to verified professionals via a moderation and notification system.

Challenges we ran into One major challenge was setting the right confidence thresholds for the AI model — balancing responsiveness with accuracy. Another challenge was designing a smooth handoff from AI to human experts without interrupting the user experience. Building a trustworthy network of professionals and ensuring fast response times was also critical.

Accomplishments that we're proud of We’re proud of building a system that respects the limits of AI while empowering human expertise. The seamless integration between AI and real experts sets Plantbook apart. We’re also proud of our growing community of plant professionals who bring unique local knowledge and experience to users worldwide.

What we learned We learned that combining AI and human expertise can create a much more reliable and user-friendly experience than relying on AI alone. We also gained valuable insights into designing escalation flows and community moderation tools that maintain quality answers.

What's next for Plantbook Next, we plan to expand our professional network globally and introduce features like image-based plant diagnostics, community voting on answers, and gamified incentives for professionals. We also aim to fine-tune our AI model further to improve confidence thresholds and reduce the need for escalation when possible.

Built With

• next