Expert Python/FastAPI Developer for High-Concurrency AI Retrieval API

We are building a highly scalable "Enterprise Secure Knowledge-Base Retrieval and Analytics Dashboard." The system routes heavy user traffic (text and streaming audio) through an LLM, strictly grounding the AI's answers in verified internal documents stored in a vector database. We already have the database schema and frontend UI planned; we need an expert backend engineer to build the connective tissue. The Tech Stack - Backend: Python (FastAPI) - AI Integration: Google Vertex AI (Gemini Flash APIs) - Database Integration: PostgreSQL (with pgvector & PgBouncer) - Infrastructure: Docker & Google Cloud Run - Auth: Firebase Authentication Key Responsibilities - Build lightning-fast, highly concurrent REST APIs using FastAPI. - Integrate Google Vertex AI to handle multi-modal inputs (streaming text and native audio). - Develop a strict Retrieval-Augmented Generation (RAG) pipeline that queries our PostgreSQL database and forces the LLM to output structured data (JSON/SVG) without hallucinating. - Package the application in a clean Dockerfile for seamless serverless deployment on Google Cloud Run. Required Experience - Proven, production-level experience with Python, FastAPI, and asynchronous (async/await) programming. - Experience connecting backends to PostgreSQL and handling connection pooling. - Direct experience working with modern LLM APIs (specifically Google Vertex AI/Gemini). - Fluent in Git/GitHub version control.