abdulhaq6920/fikra-ai

How abdulhaq6920/fikra-ai works

FIKRA AI is a multi-tenant RAG chatbot: organization members upload documents for asynchronous indexing, then ask questions and receive answers grounded in retrieved passages with validated citations. The map also includes authentication, administration, analytics, and the ingestion and retrieval layers. The README documents the end-to-end worker and infrastructure flow; source excerpts clarify key service calls. Some route and frontend internals were not sampled, so those links are limited to documented relationships.

Main entry points: router.py API router · document_tasks.py Document worker · layout.tsx Next.js application · session.py PostgreSQL and pgvector

Components 26

Web application

Next.js application frontend/src/app/layout.tsx
Chat interface frontend/src/app/(protected)/chat/page.tsx
Documents interface frontend/src/app/(protected)/documents/page.tsx
Admin and analytics frontend/src/app/(protected)/admin/page.tsx

API and identity

API router backend/app/api/router.py
Authentication routes backend/app/api/v1/auth.py
Auth service backend/app/services/auth_service.py
Auth dependencies backend/app/api/dependencies.py
Document routes backend/app/api/v1/documents.py
Chat routes backend/app/api/v1/chat.py
Admin and analytics routes backend/app/api/v1/admin.py
Insights routes backend/app/api/v1/admin_insights.py
Insights service backend/app/services/insights_service.py

Document ingestion

Document worker backend/app/workers/document_tasks.py
Extraction and chunking backend/app/services/ingestion_service.py
Local file storage backend/app/storage/local_storage.py

Grounded answers

RAG orchestration backend/app/services/rag_service.py
Hybrid retrieval backend/app/services/retrieval_service.py

Data and providers

PostgreSQL and pgvector backend/app/db/session.py
Embedding provider backend/app/providers/embeddings.py
Cross-encoder reranker backend/app/providers/reranker.py
LLM provider interface backend/app/providers/llm.py

Other components

Organization members
Groq API
Ollama service
Redis and Celery queue

Connections 31

Organization members → Next.js application: uses README.md
Next.js application → API router: calls API README.md
API router → Authentication routes: registers backend/app/api/router.py
API router → Chat routes: registers backend/app/api/router.py
API router → Document routes: registers backend/app/api/router.py
API router → Admin and analytics routes: registers backend/app/api/router.py
API router → Insights routes: registers backend/app/api/router.py
Authentication routes → Auth service: calls backend/app/api/v1/auth.py
Chat routes → RAG orchestration: calls backend/app/api/v1/chat.py
Insights routes → Insights service: calls backend/app/api/v1/admin_insights.py
Admin and analytics routes → PostgreSQL and pgvector: queries backend/app/api/v1/admin.py
Document routes → Redis and Celery queue: queues ingestion README.md
Redis and Celery queue → Document worker: dispatches tasks README.md
Document worker → Extraction and chunking: calls backend/app/workers/document_tasks.py
Document worker → Local file storage: uses backend/app/workers/document_tasks.py
Document worker → Embedding provider: embeds chunks backend/app/workers/document_tasks.py
Document worker → PostgreSQL and pgvector: persists index README.md
Chat interface → Next.js application: renders in
Documents interface → Next.js application: renders in
Admin and analytics → Next.js application: renders in
RAG orchestration → Hybrid retrieval: retrieves passages backend/app/services/rag_service.py
Hybrid retrieval → Embedding provider: embeds query backend/app/services/retrieval_service.py
Hybrid retrieval → Cross-encoder reranker: reranks results backend/app/services/retrieval_service.py
Hybrid retrieval → PostgreSQL and pgvector: searches passages backend/app/services/retrieval_service.py
RAG orchestration → LLM provider interface: generates answer backend/app/services/rag_service.py
LLM provider interface → Groq API: uses when configured README.md
LLM provider interface → Ollama service: uses when configured README.md
RAG orchestration → PostgreSQL and pgvector: loads conversation backend/app/services/rag_service.py
Authentication routes → Auth dependencies: uses dependencies backend/app/api/v1/auth.py
Chat routes → Auth dependencies: uses dependencies backend/app/api/v1/chat.py
Insights routes → PostgreSQL and pgvector: queries backend/app/api/v1/admin_insights.py

Drawn by GitDiagram on November 9, 2026 from the repository's file tree, README and source files. Read as Markdown ·