Senior AI Backend Engineer Needed to Build a Production AI Platform (RAG, LLMs, Django/FastAPI)

We're building an AI-powered platform that combines document intelligence, Retrieval-Augmented Generation (RAG), LLM workflows, and scalable backend infrastructure. This is not a simple ChatGPT wrapper or weekend AI demo. The platform is intended to support thousands of users, multiple AI providers, document processing pipelines, background workers, and production-grade APIs. We're looking for an experienced engineer who has already built and shipped real AI systems. You'll work directly with the founder to build the first production version of the platform and establish the technical foundation for future development. Important: Personalized Loom Video Required Please include a 5–10 minute Loom video created specifically for this job. Generic introduction videos will be ignored. In your Loom, we'd like you to walk through a relevant project you've built and explain: The overall architecture Your role in the project How the AI pipeline worked Which LLMs and embedding models you used How documents were processed How retrieval was implemented How background jobs were handled Any scaling or production challenges you solved What you would prioritize when building a production AI backend Please hide any confidential client information. About the Project The platform allows users to upload documents, create AI assistants, and interact with their own knowledge through natural language. Core features include: Document upload and ingestion OCR support Embedding generation Vector search RAG pipelines Conversation history AI provider integrations Background processing User authentication Admin dashboard Usage tracking API access The goal is to build a backend that is clean, scalable, and production-ready rather than simply making something that works. Core Responsibilities Design backend architecture Build scalable APIs Implement RAG pipelines Build document ingestion workflows Integrate OpenAI, Gemini, Claude, or similar models Optimize vector search performance Build background job pipelines Improve logging and monitoring Design database schemas Write clean, maintainable code Help define technical direction Current Technology Python Django / Django REST Framework FastAPI PostgreSQL Redis Celery Docker OpenAI Gemini LangChain (optional) Milvus / Qdrant / Weaviate GitHub Actions Experience Required We're looking for someone who has actually built production AI systems. Ideal experience includes: Production RAG systems Vector databases Embedding pipelines Multi-document search FastAPI or Django PostgreSQL optimization Redis and Celery Docker AI API integrations Authentication and RBAC Production deployments Nice to Have OCR Whisper Multi-modal LLMs Computer Vision Multi-agent systems Kubernetes AWS or GCP CI/CD pipelines WebSockets Multi-tenant SaaS architecture Not a Good Fit If You've only built chatbot demos. Most of your work is prompt engineering without backend ownership. You rely heavily on low-code/no-code AI tools. You haven't deployed AI systems to production. You need detailed step-by-step instructions for every task. First Milestone (Fixed Price – $120) We'll begin with a paid milestone focused on building one production-ready backend component. Depending on your expertise, this may involve: Building part of the document ingestion pipeline Implementing a RAG workflow Creating vector search APIs Integrating an LLM provider Building authentication and RBAC Setting up background workers Successful completion will lead to additional milestones and a long-term collaboration. Please Include in Your Proposal A personalized Loom video A brief introduction One production AI project you've built Your role in that project The technologies you used Which vector database you've worked with Which LLM providers you've integrated Links to your GitHub or portfolio (if available) Your estimated availability