llm-application-architecturelisted

# LLM Application Architecture ## Purpose Create an LLM application architecture that delivers actionable, measurable results. **Category**: AI & Automation ## Inputs ### Required - **Objective**: What you want to achieve with this deliverable - **Context**: Relevant background information ### Optional - **Constraints**: Any limitations or requirements to consider - **Existing Work**: Previous documents or data to build on ## Context Before starting, read the repo's `CLAUDE.md` and any relevant notes under `memory/` (e.g. `memory/repo/<repo>/`, `memory/feedback/`) for prior decisions and constraints. ## Process ### Step 1: Context & Research - Review any existing llm application architecture documents in the project - Identify key stakeholders and their requirements - Select the most appropriate framework: Anthropic 6 Agent Patterns, RAG Architecture, GenAI Platform Stack ### Step 2: Analysis & Framework Application - Apply the selected framework to structure the llm application architecture - Identify gaps, opportunities, and risks - Define success metrics: TTFT (Time to First Token), Retrieval Precision/Recall, Guardrail Trigger Rate, End-to-End Latency - Document assumptions and dependencies - Validate approach against industry best practices ### Step 3: Build the Deliverable - Structure the llm application architecture using the output format below - Include specific, actionable recommendations — not generic advice - Add concrete numbers, timelines, and benchm