Python/FastAPI + LangGraph engineer to take an LLM product to beta launch

I'm a founder building an early-stage software product that uses LLMs to help professional services teams with complex written work. The backend is built and has been through an independent security review. I'm now looking for an engineer to take it through hardening, deployment and an invite-only beta. THE STACK Python, FastAPI, LangGraph, PostgreSQL with pgvector, Alembic, Django (admin), S3-compatible object storage, Anthropic and OpenAI APIs, LangSmith. WHAT'S NEXT - Security hardening from the review: rate limiting, auth and tokens, tenant isolation - Updating the model integration to newer LLM versions and managing cost per run - Production deployment to the cloud - Connecting the front end for beta users - Ongoing features after launch WHAT I'M LOOKING FOR - 2–5 years building production backends in Python - Hands-on experience shipping LLM apps: agent workflows, RAG, prompt and cost management - Solid understanding of multi-tenant data security - Clear written communication. You'll be working with a non-technical founder, so explaining trade-offs plainly matters. - Nice to have: cloud deployment (AWS, GCP or similar), React, Stripe HOW IT WORKS - Ongoing hourly contract, around 15–25 hours a week to start - Some overlap with Australian business hours (AEST) - A senior engineer reviews architecture and security - Starts with a short paid trial task - NDA and IP assignment required before access to the codebase Independent freelancers only, please, no agencies. To show you've read this, start your proposal with the word "pgvector".