Job description
What this role is about
We are looking for an experienced AI Full-Stack Developer to build and launch a production-ready web application with a RAG-powered chatbot and a modern Next.js frontend. Project Overview: We are building an AI assistant that can answer user questions based on our own documents and knowledge base (not just general LLM responses). The application should provide accurate, source-grounded answers, clean UX, and scalable architecture for future growth. Scope of Work: Build a RAG chatbot pipeline (ingestion, chunking, embeddings, retrieval, generation). Implement document ingestion for common formats (PDF, DOCX, TXT, web pages). Create a robust retrieval strategy (metadata filters, top-k, reranking optional). Integrate LLM APIs and prompt orchestration. Build a polished Next.js frontend chat interface. Implement authentication and user/session handling. Add conversation history and basic admin controls. Add observability (logging, error handling, analytics). Support deployment and environment setup.
Requirements
Strong experience with RAG architecture in production. Strong Next.js (App Router preferred), React, TypeScript. Experience with vector databases (e.g., Pinecone, Weaviate, pgvector, Chroma). Experience with LLM providers (OpenAI/Anthropic/etc.). Backend API design (Node.js/TypeScript preferred). Prompt engineering and evaluation for answer quality. Familiarity with cloud deployment (Vercel/GCP/AWS).