LLM / RAG Engineer – Python, Pinecone, LangChain

Role Overview

We are seeking a senior Python/AI developer with proven experience in Retrieval-Augmented Generation (RAG) systems to build a production-ready integration between a Pinecone vector database and ChatGPT. The goal is to create a clean, scalable, and highly user-friendly foundation that allows non-technical users to easily upload large volumes of documents (articles, books, PDFs, etc.) and benefit from enhanced, context-aware ChatGPT responses.

Responsibilities

• Configure and optimize a Pinecone vector database for scalability and performance.
• Integrate Pinecone with ChatGPT via a RAG pipeline, leveraging frameworks such as LangChain or LlamaIndex.
• Implement efficient chunking and embedding strategies to support hundreds (or thousands) of uploaded documents.
• Design intuitive workflows that let non-coders easily upload and manage documents (e.g., drag-and-drop upload, simple web interface, or lightweight dashboard).
• Ensure ChatGPT responses are enriched with context from the uploaded knowledge base.
• Provide clear, professional documentation and a brief training/hand-off session for ongoing use and maintenance.

Required Skills

• Strong Python development background with focus on AI/ML integrations.
• Direct experience with Pinecone or other vector databases.
• Expertise building RAG pipelines using LangChain, LlamaIndex, or similar frameworks.
• Familiarity with the OpenAI API and best practices for LLM integration.

Nice to Have

• Experience designing user-facing tools (simple UIs, dashboards, or document upload workflows).
• Prior involvement in production-scale RAG systems handling large volumes of documents.
• Knowledge of HuggingFace embeddings or other model providers.
• Strong documentation and communication skills, with the ability to make complex systems usable by non-technical end users.

Deliverables

• A production-ready RAG pipeline integrating Pinecone and ChatGPT.
• A simple, intuitive interface (web-based or lightweight dashboard) enabling non-technical users to:
  Upload and manage documents (hundreds of articles/books).
  Trigger re-indexing automatically.
  Query ChatGPT with enriched, context-aware responses from the uploaded knowledge base.
• Documentation and instructions for ongoing maintenance and scaling.