How arunchandrasekar07/devmind works
DevMind is a repository-aware AI assistant. The main workflow is project setup and repository upload, followed by parsing, chunking, embedding and indexing; users then ask project-specific questions and receive context-grounded AI responses. Authentication, project management and analytics are supporting product stages. The graph reflects documented frontend-to-API flow and source-indexed backend relationships; frontend wiring and some backend internals were not excerpted, so uncertain links are omitted.
Main entry points: main.py FastAPI Application · pipeline.py Ingestion Pipeline · project_service.py Project Service · app Next.js Interface
Components 30
User Experience
- Next.js Interface
frontend/src/app - Project Workspace
frontend/src/app/dashboard/projects/[id]/page.tsx - Repository Upload
frontend/src/components/project/FileUploadZone.tsx - Chat Interface
frontend/src/app/dashboard/projects/[id]/chat/page.tsx - Voice Input
frontend/src/lib/hooks/useVoiceInput.ts - Analytics Dashboard
frontend/src/app/dashboard/projects/[id]/analytics/page.tsx - API Client
frontend/src/lib/api.ts
API and Identity
- FastAPI Application
backend/app/main.py - Authentication Routes
backend/app/routers/auth.py - Authentication Service
backend/app/services/auth_service.py - Token Authentication
backend/app/middleware/auth.py - Project Routes
backend/app/routers/projects.py - Document Routes
backend/app/routers/documents.py - Chat Routes
backend/app/routers/chat.py - Project Service
backend/app/services/project_service.py
Repository Indexing
- Ingestion Pipeline
backend/app/services/ingestion/pipeline.py - File Parsing
backend/app/services/ingestion/file_parser.py - Code Chunking
backend/app/services/ingestion/chunker.py
AI Answering
- Embedding Generation
backend/app/services/ai/embedder.py - RAG Orchestrator
backend/app/services/ai/rag_pipeline.py - Semantic Retrieval
backend/app/services/ai/retriever.py - Multi-Provider LLM
backend/app/services/ai/llm_client.py
Persistent State
- Database Sessions
backend/app/database.py - Repository and App Data
backend/app/models/document.py - User Records
backend/app/models/user.py - Project Records
backend/app/models/project.py - Conversation Records
backend/app/models/chat.py
Other components
- Developer
- Supabase Auth
- LLM Providers
Connections 36
- Developer → Next.js Interface: uses
README.md - Next.js Interface → FastAPI Application: calls API
README.md - Next.js Interface → Project Workspace: opens workspace
- Project Workspace → Repository Upload: offers upload
- Project Workspace → Chat Interface: opens chat
- Chat Interface → Voice Input: accepts voice
- Next.js Interface → Analytics Dashboard: shows analytics
- API Client → FastAPI Application: sends requests
- FastAPI Application → Authentication Routes: routes auth
backend/app/main.py - FastAPI Application → Project Routes: routes projects
backend/app/main.py - FastAPI Application → Document Routes: routes documents
backend/app/main.py - FastAPI Application → Chat Routes: routes chat
backend/app/main.py - Authentication Routes → Authentication Service: authenticates users
backend/app/routers/auth.py - Authentication Routes → Token Authentication: requires identity
backend/app/routers/auth.py - Token Authentication → Authentication Service: validates tokens
backend/app/middleware/auth.py - Authentication Service → Supabase Auth: manages credentials
backend/app/services/auth_service.py - Project Routes → Project Service: manages projects
backend/app/routers/projects.py - Document Routes → Ingestion Pipeline: ingests files
backend/app/routers/documents.py - Document Routes → Project Service: updates projects
backend/app/routers/documents.py - Chat Routes → RAG Orchestrator: queries context
backend/app/routers/chat.py - Chat Routes → Project Service: loads project
backend/app/routers/chat.py - Ingestion Pipeline → File Parsing: parses files
backend/app/services/ingestion/pipeline.py - Ingestion Pipeline → Code Chunking: chunks code
backend/app/services/ingestion/pipeline.py - Ingestion Pipeline → Embedding Generation: embeds chunks
backend/app/services/ingestion/pipeline.py - RAG Orchestrator → Semantic Retrieval: retrieves context
backend/app/services/ai/rag_pipeline.py - RAG Orchestrator → Multi-Provider LLM: generates answers
backend/app/services/ai/rag_pipeline.py - Semantic Retrieval → Embedding Generation: embeds queries
backend/app/services/ai/retriever.py - Multi-Provider LLM → LLM Providers: routes generation
README.md - Project Service → Project Records: reads and writes
backend/app/services/project_service.py - Project Service → Repository and App Data: updates counters
backend/app/services/ingestion/pipeline.py - Ingestion Pipeline → Repository and App Data: stores chunks
backend/app/services/ingestion/pipeline.py - Database Sessions → Repository and App Data: backs models
backend/app/models/document.py - Database Sessions → User Records: backs models
backend/app/models/user.py - Database Sessions → Project Records: backs models
backend/app/models/project.py - Chat Routes → Conversation Records: uses conversations
backend/app/routers/chat.py - Authentication Service → User Records: syncs profiles
backend/app/services/auth_service.py