AI/RAG Engineer for SaaS Audit

I’m looking for an experienced **AI/RAG Engineer** to review an existing AI-powered SaaS MVP that I built using Lovable. The MVP is already functional. I need an **independent technical audit before continuing development**, with the main focus on: **1. Answer quality — highest priority** * RAG architecture and retrieval quality * Accuracy and grounding of AI answers * Use of authoritative and up-to-date sources * Embeddings, chunking, ranking/re-ranking * Prompt architecture and hallucination prevention * Citations and source traceability * How user-specific information affects answers * Analyse and Draft a Letter functionality **2. Speed — second priority** * RAG/database/vector-search performance * API and LLM latency * Unnecessary API calls/processes * Caching, indexing, parallelisation and streaming opportunities **3. Regulatory monitoring** The app is designed to monitor UK regulatory sources, detect changes, automatically ingest and re-index updated information, and provide personalised alerts. I need you to verify whether this pipeline actually works, including monitoring jobs, RSS/content-change detection, ingestion, indexing, RAG retrieval and personalised alerts. **4. Security & scalability** Review the current architecture, Supabase/database security, authentication, user-data isolation, API security and scalability. ### Deliverable I want a clear technical audit showing: * What is working well * Critical problems * What should be **Kept / Improved / Rebuilt / Not Built Yet** * P0/P1/P2/P3 priorities * Top recommended changes * Estimated engineering effort **Important:** This is an audit first. Please do not rebuild or substantially modify the application during the initial review. If the audit is successful, there may be a separate follow-up contract to implement the agreed changes. I am particularly interested in engineers with proven experience in **production RAG systems, LLM applications, vector databases, data pipelines, AI evaluation and performance optimisation**. Please include relevant examples of your previous work in your proposal.