AI Real Estate Assistant is an AI-powered platform designed for conversational property search, financial analytics, and market insights. It targets individuals and professionals seeking a streamlined, data-driven approach to real estate, offering benefits like natural language property discovery and comprehensive financial planning tools.

How It Works

The platform employs a hybrid architecture featuring a FastAPI backend and a Next.js frontend, leveraging ChromaDB for efficient semantic and keyword property search. Its core innovation lies in a sophisticated AI engine that supports over six LLM providers (including OpenAI, Anthropic, and Google) with intelligent routing and automatic fallback mechanisms for reliability. Queries are processed by a Query Analyzer, which directs them to either a Retrieval Augmented Generation (RAG) engine for simpler requests or a Hybrid Agent for complex ones, utilizing tools and vector databases for enhanced relevance and accuracy.

Quick Start & Requirements

The recommended setup involves Docker, with PowerShell scripts provided for Windows users to launch containers, generate demo data, and manage the demo environment. Manual Docker Compose setup and direct backend/frontend installations are also supported. The project includes over 250 properties across five Polish cities for local demo data. Production deployment necessitates API keys for chosen LLM providers. Links to a live demo, local setup documentation, and a 5-minute quickstart guide are available.

Highlighted Details

• Multi-Provider AI: Integrates 6+ LLM providers with intelligent routing and fallback chains.
• Smart Property Search: Utilizes natural language processing with ChromaDB and MMR reranking for hybrid semantic and keyword search, claiming 30-40% better relevance.
• Financial Tools: Offers mortgage calculation, rent-vs-buy comparisons, ROI analysis, and Comparative Market Analysis (CMA) reports.
• Multilingual Support: Features 9 languages with EU AI Act compliance labels.
• Enterprise Security: Implements OWASP-hardened measures including rate limiting, audit logging, and a progressive 5-stage security pipeline.
• Extensive Testing: Boasts over 7,000 tests (6,254 backend, 1,000 frontend) with high coverage.

Maintenance & Community

The project has 6 contributors and shows active development with over 1177 commits and extensive test coverage. Discussions and feature planning are primarily managed through GitHub Issues.

Licensing & Compatibility

The project is released under the MIT License, which is permissive and allows for commercial use and integration into closed-source applications.

Limitations & Caveats

The live demo utilizes simulated AI responses; production use requires configuring API keys for LLM providers. Deployments on free-tier cloud services like Render may experience cold starts after periods of inactivity. Advanced features such as multi-tenant architecture and a billing API are listed on the roadmap, indicating they are not yet implemented.