AI Engineer Needed to Build Agentic AI Shopping Assistant (LLM, RAG, MCP, Python)

Project Overview We are looking for an experienced AI Engineer to help us build an intelligent AI-powered shopping assistant for an e-commerce platform. The goal is to create an AI assistant that can understand customer queries, recommend relevant products, answer product-related questions, and perform shopping-related actions through intelligent tool calling. We want to go beyond a traditional chatbot and build an agentic AI system that can understand user intent, maintain conversational context, make decisions, and execute actions using backend tools. Key Responsibilities Design and develop AI agents using modern LLM frameworks and architectures. Build RAG pipelines for retrieving relevant information from product catalogs, documentation, and knowledge bases. Implement semantic search, embeddings, vector databases, and hybrid retrieval strategies. Develop intelligent tool-calling workflows for product discovery, filtering, recommendations, and cart operations. Build MCP servers to connect AI agents with external systems and APIs. Implement conversational memory and context management for multi-turn conversations. Integrate LLM providers such as OpenAI, Anthropic, or other suitable models. Develop scalable backend services using Python and FastAPI. Implement AI observability, tracing, and evaluation using tools such as Langfuse or LangSmith. Optimize AI workflows for response quality, latency, and token costs. Required Technical Skills Strong experience with Python and FastAPI. Hands-on experience building Agentic AI applications. Solid understanding of LLMs, prompt engineering, and function/tool calling. Practical experience with RAG architecture and retrieval pipelines. Experience with embedding models and vector databases. Understanding of MCP architecture and MCP server development. Experience integrating third-party APIs and backend services. Knowledge of conversational memory and multi-agent workflows. Nice to Have Experience with LangChain or LangGraph. Experience with LiteLLM. Experience with Langfuse or LangSmith. Experience with PostgreSQL, Redis, or other databases. Experience building AI-powered e-commerce applications. Experience with Next.js or React. Expected Deliverables Working AI shopping assistant with conversational capabilities. RAG pipeline for product and knowledge retrieval. Agentic workflows with backend tool integrations. MCP server implementation for external system connectivity. Conversation history and memory management. Basic evaluation and observability setup. Clean, maintainable, and well-documented code. What We're Looking For We are looking for someone who has actually built and deployed AI-powered applications, not someone who has only experimented with LLM APIs or created basic prompt-based chatbots. You should understand how to design reliable AI systems where the LLM handles reasoning and intent understanding while backend services handle business logic, data retrieval, and actual operations. We value clean architecture, practical problem-solving, and the ability to make the right technical decisions. How to Apply Please answer the following questions in your proposal: Describe an Agentic AI application you have built. What was the architecture and which technologies did you use? Have you implemented a production-ready RAG pipeline? Explain your retrieval strategy, embedding model, and vector database. Have you developed custom MCP servers? If yes, share an example. How do you manage conversational memory and context in multi-turn AI applications? How do you prevent LLMs from executing incorrect tools or making unreliable decisions? Share links or brief descriptions of relevant AI projects you have worked on. We are interested in working with an engineer who can contribute to the architecture and development of a reliable, production-ready AI application. Long-term collaboration is possible for the right candidate.