About This Architecture

Ani Enterprise GenAI Search Platform on Azure integrates user channels through Azure Front Door and WAF, routing authenticated requests via Keycloak and Azure Entra ID to a modular backend. The Ani Platform orchestrates search, chat, and RAG workflows across LiteLLM Gateway, embedding services, and PostgreSQL vector databases, while Content Processing pipelines ingest enterprise data from productivity, collaboration, and file systems. Comprehensive monitoring via logging, metrics, tracing, and alerting ensures observability across all tiers. This architecture demonstrates enterprise-grade GenAI at scale with security, identity governance, and knowledge management as first-class concerns. Fork and customize this diagram on Diagrams.so to adapt connectors, LLM providers, or data sources to your organization's needs.

People also ask

How do you build a secure, scalable enterprise GenAI search platform on Azure with RAG and vector embeddings?

The Ani Enterprise GenAI Search Platform on Azure layers security (Azure Front Door, WAF, DDoS), identity (Keycloak, Entra ID), and a modular backend: Ani Platform orchestrates search and chat via LiteLLM Gateway and embedding services; Knowledge Platform stores vectors in PostgreSQL and graphs in object storage; Content Processing ingests from enterprise sources. Monitoring and observability span