Senior Backend / Cloud Engineer (FastAPI, PostgreSQL/pgvector, AWS, LangGraph)

I am the founder of a funded EdTech/Enterprise Compliance startup building "Project Vox," a low-latency, voice-to-voice AI evaluation engine. I am NOT looking for someone to build from scratch. I already have the foundational architecture, the PostgreSQL database schema, the LangGraph logic, and the highly complex LLM system prompts written. I need a Senior Cloud/Backend Engineer to take my code, deploy it securely, and build the API bridge for our mobile frontend. Your Specific Tasks: Cloud Provisioning & Security: Deploy a Managed PostgreSQL database (with the pgvector extension) on AWS or GCP. You must configure a VPC so the database has ZERO public internet access (bank-level security). API Wrapping: Wrap my existing LangGraph Python scripts into a scalable FastAPI server. API Integrations: Securely connect our enterprise LLM API (Claude/Gemini) and our Speech-to-Text/Text-to-Speech engines (Deepgram) via WebRTC or WebSockets for sub-second latency. Secret Management: Ensure zero hardcoded API keys by utilizing AWS Secrets Manager or GCP Secret Manager. Requirements: 5+ years of Senior Backend experience (Python, Node, or Go). Deep expertise in AWS or GCP cloud security architecture. Experience with Vector Databases (pgvector, Pinecone, etc.). Experience handling live audio streaming or WebSockets is a massive plus. Screening Questions (Please answer to be considered): How exactly would you configure the AWS VPC to ensure the pgvector database is completely hidden from the public internet, but accessible to the FastAPI server? How do you handle secrets and API keys in production? Have you ever deployed an app using WebRTC or WebSockets for live streaming?