AI/ML Code & RAG Pipeline Review with Actionable Report

We are looking for an experienced AI/ML engineer to review our existing AI/ML codebase, RAG pipeline, and/or LLM integrations and provide a thorough technical assessment. The goal is to identify hidden inefficiencies, architectural weaknesses, robustness issues, and potential production risks that may not be obvious during normal development. What we need you to review Depending on the scope of the project, this may include: 1. Architecture & design — retrieval logic, chunking strategy, embedding/retrieval approach, prompt structure, LLM orchestration, and agent workflows 2. Performance & efficiency — unnecessary LLM calls, inefficient database/vector queries, token usage, latency, and potential scaling bottlenecks 3. Robustness & reliability — error handling, edge cases, failure modes, hallucination risks, prompt injection, retrieval poisoning, and other security concerns 4. Code quality & maintainability — code structure, separation of concerns, duplication, readability, and areas that may become difficult to maintain or extend 5. AI/ML implementation — whether the current approach is technically sound and appropriate for the problem 6. Best practices — identify areas where our implementation differs from current industry best practices and recommend practical improvements Deliverables We expect a written technical review/report containing: * Specific issues with file and line references where possible * A clear explanation of why each issue matters * Prioritized recommendations (Critical → High → Medium → Nice-to-have) * Suggested approaches for fixing the most important issues * Identification of potential production/scaling risks * Clear next steps and recommendations for improving the system We are not looking for a generic AI code review or a list of theoretical best practices. We want someone who can understand the existing architecture, trace how the components work together, identify real weaknesses, and provide actionable recommendations. Ideal candidate You should have strong practical experience with: * LLM applications and production AI systems * RAG pipelines and retrieval systems * Vector databases / semantic search * Prompt engineering and LLM orchestration * AI/ML system architecture * Python and backend development * Performance and scalability analysis * Production reliability and AI security Experience reviewing existing production or near-production AI systems is strongly preferred.