RecoAgent

• 

  signup/login

• 

• 

• 

• 

• 

Inspiration

RecoAgent was inspired by the need for easy-to-use, powerful AI chat solutions for businesses. We aimed to make sophisticated customer engagement accessible without complex coding, bridging the gap between simple chatbots and resource-intensive support.

What it does

RecoAgent is a SaaS platform for creating, deploying, and managing AI chat agents. It allows users to build custom agents, enhance content with AI, embed agents on any website, manage conversations, and track performance through analytics. A guest mode is available for testing.

How we built it

The frontend uses React, Vite, Tailwind CSS, and Lucide React. The backend and database are powered by Supabase (PostgreSQL, Auth, RLS, Edge Functions). AI integration is handled via Groq's API, with calls securely routed through Supabase Edge Functions. The embeddable chat widget is a custom vanilla JavaScript file that loads the React component. The application is deployed on Netlify.

Challenges we ran into

Key challenges included implementing robust Supabase Row Level Security for both authenticated and anonymous users, securely integrating the AI API via serverless functions, designing a flexible and customizable embeddable widget, managing real-time conversation states, and creating a functional guest mode.

Accomplishments that we're proud of

We are proud of the seamless end-to-end user experience, the robust security model achieved through Supabase RLS and secure Edge Functions, the highly customizable chat widget, and the AI-powered assistance for agent creation. The scalable architecture is also a significant accomplishment.

What we learned

We gained deep insights into Supabase's RLS and Edge Functions, the importance of clear API contracts, and best practices for building embeddable JavaScript components. We also learned about the nuances of prompt engineering and the value of a user-centric design approach.

What's next for RecoAgent

Future plans include advanced analytics, expanding the knowledge base capabilities, integrating with CRM systems, adding voice functionalities, and potentially offering a wider range of AI models through a marketplace.

Built With

• date-fns
• framer
• groq
• lucide
• netlify
• react
• recharts
• supabase
• tailwind
• typescript
• vite