How priyanshtech/konvert-ai- works
ai-chat is a multi-tenant AI support product: teams manage bots and knowledge, while visitors chat through an embeddable widget. The backend routes authentication, bot management, knowledge ingestion, chat, and reporting; ingestion parses content and creates embeddings, while chat retrieves relevant chunks and streams generated answers. The frontend includes bot, knowledge, conversation, lead, analytics, and simulator screens. README provides no further product detail. Some UI-to-API wiring and unsampled subsystem internals are not evidenced, so those links are omitted.
Main entry points: main.py API application · ingestion_worker.py Ingestion worker · models.py Domain models · auth.py Authentication
Components 21
Web and Widget
- Team workspace
frontend/src/app - Embeddable widget
frontend/public/widget.js
Backend API
- API application
backend/app/main.py - Authentication
backend/app/api/auth.py - Bot management
backend/app/api/bots.py - Chat endpoint
backend/app/api/chat.py - Reporting API
backend/app/api/dashboard.py - Knowledge API
backend/app/api/knowledge.py - Widget API
backend/app/api/widget.py
Knowledge and Retrieval
- Document parsing
backend/app/services/document_parser.py - Ingestion worker
backend/app/workers/ingestion_worker.py - Embedding service
backend/app/services/embedding.py - RAG engine
backend/app/services/rag_engine.py
Identity and Data
- Access security
backend/app/core/security.py - Domain models
backend/app/models/models.py - Application database
backend/app/core/database.py - Runtime configuration
backend/app/core/config.py
Other components
- Support team
- Chat visitor
- Redis queue
- AI provider
Connections 27
- API application → Authentication: registers
backend/app/main.py - API application → Bot management: registers
backend/app/main.py - API application → Chat endpoint: registers
backend/app/main.py - API application → Reporting API: registers
backend/app/main.py - API application → Knowledge API: registers
backend/app/main.py - API application → Widget API: registers
backend/app/main.py - Authentication → Access security: uses
backend/app/api/auth.py - Authentication → Domain models: queries
backend/app/api/auth.py - Authentication → Application database: uses
backend/app/api/auth.py - Bot management → Access security: uses
backend/app/api/bots.py - Bot management → Domain models: queries
backend/app/api/bots.py - Bot management → Application database: uses
backend/app/api/bots.py - Chat endpoint → RAG engine: invokes
backend/app/api/chat.py - Chat endpoint → Access security: uses
backend/app/api/chat.py - Chat endpoint → Domain models: queries
backend/app/api/chat.py - Reporting API → Access security: uses
backend/app/api/dashboard.py - Reporting API → Domain models: queries
backend/app/api/dashboard.py - Knowledge API → Ingestion worker: queues jobs
backend/app/api/knowledge.py - Ingestion worker → Document parsing: parses content
backend/app/workers/ingestion_worker.py - Ingestion worker → Embedding service: embeds chunks
backend/app/workers/ingestion_worker.py - Ingestion worker → Domain models: writes records
backend/app/workers/ingestion_worker.py - Ingestion worker → Redis queue: enqueues jobs
backend/app/workers/ingestion_worker.py - RAG engine → Embedding service: embeds queries
backend/app/services/rag_engine.py - RAG engine → AI provider: requests completions
backend/app/services/rag_engine.py - Embedding service → AI provider: requests vectors
backend/app/services/embedding.py - Support team → Team workspace: uses
- Chat visitor → Embeddable widget: chats through