Back to jobs

AI Customer Support Agent with RAG Knowledge Base & Admin Dashboard

Search - AI Chatbot · local_filter_skipped · UID ~022077321104282126940

Open Job

Job Details

Budget Unknown
ExperienceExpert
DurationUnknown
Weekly hoursUnknown
Client countryUnknown
ProposalsUnknown
InterviewingUnknown
Invites sentUnknown
First seenWed, Jul 15, 2026 12:30 PM
Last seenWed, Jul 15, 2026 8:19 PM

Description

Project Description I am looking for a Full Stack AI Engineer to build an AI-powered Customer Support Assistant that can answer customer questions using company documents and knowledge base. The application should have a modern React.js frontend and a FastAPI backend with PostgreSQL database. The AI assistant will use Retrieval-Augmented Generation (RAG) to provide accurate responses from uploaded PDF and DOCX files instead of generating random answers. The system should include secure authentication, conversation history, document management, and an admin dashboard where administrators can upload new documents and monitor AI interactions. The chatbot should provide fast, reliable, and context-aware answers while maintaining a clean and responsive user interface. The project should be production-ready, scalable, and designed with clean architecture so it can be extended with future AI features such as AI agents, workflow automation, and CRM integration. Key Requirements - Build a responsive frontend using React.js, Next.js, and Tailwind CSS. - Develop the backend using Python FastAPI with PostgreSQL. - Implement secure authentication and user management. - Build REST APIs for all frontend operations. - Create a document upload system supporting PDF and DOCX files. - Generate embeddings and implement a RAG pipeline for document retrieval. - Integrate an LLM to answer questions based only on uploaded documents. - Store chat history and document metadata. - Create an Admin Dashb

Skills

AI Agent Development Python React LangChain

Notification History

ChannelTypeStatusSentError
No notifications.

User Actions

ActionActed at
No actions.