AI Full-Stack Developer Needed for RAG-Based AI Assistant Platform

We are looking for an experienced AI Full-Stack Developer to help us build an AI-powered knowledge assistant that allows users to upload or connect business data and interact with it through a conversational interface. The application will combine LLMs, Retrieval-Augmented Generation (RAG), backend APIs, vector search, and a modern frontend dashboard. The goal is to build a clean and extensible AI application where users can manage knowledge sources, ask questions, view AI-generated answers with supporting context, and track conversations and AI executions. What We Need We need someone who can help design and implement the full-stack AI application, including: AI-powered chat interface RAG-based question answering Document upload and ingestion Text extraction, cleaning, and chunking Embedding generation and vector storage Semantic search and retrieval LLM integration using OpenAI, Claude, AWS Bedrock, or similar Conversation and message history Backend APIs for chat and knowledge management User authentication and authorization Knowledge-source management dashboard AI response citations or source references Loading, streaming, and error states on the frontend Database models for users, conversations, documents, and AI executions Logging and basic AI observability Clean architecture allowing additional AI agents/features to be added later AI Workflow A typical request may follow this flow: User Question → Backend API → Query Understanding → Vector Search → Relevant Context Retrieval → LLM / AI Agent → Response Generation → Source Attribution → Conversation Storage → Frontend Response For more advanced queries, the system may also support tool-based or agentic workflows where the AI can decide whether it needs to retrieve documents, call an API, or perform another backend operation before generating the final response. Frontend The frontend should provide a clean dashboard where users can: Chat with the AI assistant View conversation history Upload and manage documents View indexed knowledge sources See referenced sources used in AI answers Track document processing status View basic AI execution information Configure selected AI settings where appropriate Preferred frontend technologies include: React Next.js TypeScript Tailwind CSS or similar Backend The backend should expose clean APIs for: Chat and AI execution Conversation management Document upload Document ingestion Knowledge-source management Semantic retrieval User management AI execution status Error handling Preferred technologies include: Python / FastAPI Node.js / TypeScript PostgreSQL MongoDB Redis where required AI / RAG Components The system may use technologies such as: OpenAI Anthropic Claude AWS Bedrock LangChain or LlamaIndex where appropriate pgvector, Pinecone, OpenSearch, Qdrant, or similar vector stores Reranking models Structured outputs / tool calling Agentic workflows We are open to architecture recommendations from the developer. Cloud / Deployment Experience with AWS is preferred, particularly: AWS Bedrock S3 ECS Lambda API Gateway EventBridge CloudWatch RDS / PostgreSQL Docker-based deployment experience is also preferred. Expected Deliverables Recommended AI application architecture Working full-stack AI assistant RAG ingestion and retrieval pipeline Backend APIs Frontend chat/dashboard Document management functionality Conversation persistence Vector search integration LLM integration Source/citation support Error handling and logging Basic deployment configuration Documentation explaining architecture and how additional AI functionality can be added Ideal Candidate The ideal developer has experience with: Full-stack application development LLM applications Retrieval-Augmented Generation AI agents and tool calling Python / FastAPI or Node.js React / Next.js PostgreSQL / MongoDB Vector databases AWS Docker REST APIs AI observability and evaluation Production AI application architecture Experience with multi-tenant SaaS applications or AI workflow/orchestration systems is a plus. Goal The goal is to build a solid foundation for an AI-native SaaS application rather than just a basic chatbot. We want an architecture where additional capabilities such as AI agents, external API integrations, automated workflows, advanced RAG, evaluation pipelines, and business-specific AI tools can be added over time without rebuilding the entire system. Please apply with examples of AI, RAG, agentic, or full-stack AI applications you have previously built. Also briefly explain how you would structure the frontend, backend, RAG pipeline, vector storage, and LLM integration for this project.