A full-stack, AI-powered SaaS platform providing dynamic, personalized recommendations for any website or application. Built for the AI SUMMIT - 2025.
NexusAI solves the "paradox of choice" for modern e-commerce and content platforms. Instead of building a complex, in-house recommendation engine, businesses can hire our AI Agent via a simple API. Our agent provides intelligent, real-time recommendations that are controllable via a developer-friendly dashboard, ensuring that the AI's suggestions are always aligned with the business's strategic goals.
- Dynamic AI Reasoning: Utilizes a powerful LLM via LangChain to provide nuanced, personalized recommendations that adapt in real-time to user behavior.
- Explainable AI (XAI): Delivers the "why" behind each recommendation, building user trust and increasing engagement.
- Developer-Controlled Business Rules: A full-featured dashboard allows developers to "coach" the AI, creating rules to boost or demote products based on categories, tags, or other business logic.
- Dynamic Catalog Management: Developers can sync their product catalogs at any time, ensuring the AI's knowledge is always current with their inventory.
- Secure Multi-Tenant Authentication: A complete user signup, login, and dynamic API key generation system for developer accounts.
- High-Speed Session Memory: Uses Redis to maintain a short-term memory of user interactions for instantaneous personalization.
This project is a monorepo containing three distinct but interconnected applications:
packages/api-backend: The core of the service. A Python/FastAPI application that handles user authentication, data persistence, and houses the AI agent logic.packages/platform-frontend: The SaaS dashboard. A Next.js application where developers sign up, get their API keys, manage their catalogs, and define business rules.packages/demo-site: A showcase e-commerce store. A separate Next.js application that integrates the NexusAI agent, demonstrating its functionality from an end-user's perspective.
| Category | Technologies |
|---|---|
| Frontend Development | Next.js, React, TypeScript, Tailwind CSS, shadcn/ui |
| Backend Development | Python, FastAPI, SQLAlchemy |
| AI & Machine Learning | LangChain, Google Gemini, Sentence-Transformers (for Cold Start) |
| Database & Caching | PostgreSQL (hosted on Railway), Redis (via Docker) |
| Deployment | Vercel (Frontends), Render (Backend) |
Follow these instructions to set up and run the entire project on your local machine.
- Node.js (v18 or later)
- Python 3.10 and Conda (or a virtual environment manager)
- Docker Desktop (for running Redis)
- Git
git clone [https://github.com/your-username/Recommendation_Engine.git](https://github.com/your-username/Recommendation_Engine.git)
cd Recommendation_Engine