AI Engineer Needed – Complex Multi-Tenant RAG Isolation Problem

We're building a multi-tenant RAG (Retrieval-Augmented Generation) product in-house and have most of the platform up and running — ingestion, embeddings, and a working retrieval pipeline are already in place. Where we're stuck is a genuinely hard problem: guaranteeing strict, provable tenant isolation in retrieval as we scale to more customers, without tanking latency or blowing up infra cost. This is not a beginner task — we need a senior AI/ML engineer who has actually solved this class of problem before and can come in, diagnose the architecture, and fix it properly. The Problem Our current setup risks cross-tenant data leakage under certain query/retrieval conditions, and we need someone who can pinpoint exactly where and why We need an expert opinion (and implementation) on the right isolation strategy for our scale — namespace-per-tenant, metadata filtering, or separate indexes — with a clear tradeoff analysis, not just a textbook answer Retrieval accuracy and latency need to hold up even with hard isolation boundaries in place Any fix needs to be production-safe — this touches live customer data What You'll Do Audit our existing multi-tenant RAG architecture and identify isolation gaps Recommend and implement the correct isolation approach for our vector DB setup Stress-test retrieval to confirm zero cross-tenant leakage Tune for latency/cost so the fix doesn't degrade performance at scale Clearly document the root cause and the fix so our internal team can maintain it Requirements Proven, hands-on experience solving tenant isolation problems in production RAG systems (please link GitHub or case studies — no tutorials/POCs) Deep understanding of vector DB internals (Pinecone / Weaviate / Chroma / Qdrant) and multi-tenancy patterns Strong Python backend skills (FastAPI or Django) Comfortable diagnosing someone else's existing system quickly, not just building from scratch Clear communicator — you'll need to explain the root cause and tradeoffs, not just ship a patch Nice to Have Experience with LLM evaluation tooling (Ragas, LangSmith) to validate the fix Experience with tenant-aware auth/RBAC at the API gateway level Prior work in a SaaS environment with many concurrent customer orgs Engagement Details To Apply Please include: A specific example of a tenant isolation problem you've diagnosed and fixed in a past RAG project Links to relevant GitHub repos or case studies One question you'd want answered before starting this project (Generic, copy-pasted proposals will be deprioritized — we need someone who's actually solved this exact problem before.)