A premier conversational AI SaaS platform designed for high-performance reasoning and real-time web search capabilities, with subscription-based billing and multi-tenant architecture.
Qlaude operates on a modern, decoupled SaaS architecture with a frontend web application for user interactions and a separate, high-performance API backend handling AI inference, search integrations, user authentication, and billing.
The frontend client serves the customer-facing experience on its own port, integrating with Google OAuth for identity and Stripe Checkout for subscription upgrades, while seamlessly communicating with the core API for chat streaming. The backend microservice powers the platform's intelligence — multi-tenant session lifecycle, subscription quota enforcement, real-time web search orchestration via the Search Agent, and streaming chat completions through Server-Sent Events.
When a user sends a message, it's first validated against their active subscription tier. If search is enabled and authorized, the search agent queries external sources before inference. The API then streams real-time reasoning and final responses back to the client. All turns and usage metrics are securely recorded in the data layer, which manages multiple SQLite databases optimized for concurrent workloads.
The architecture follows microservice principles — the frontend client and core API scale independently, conversations are isolated for data privacy, and inference requests can be routed to any compatible LLM backend depending on load and tier requirements.
