This enhanced prompt is designed to instruct an AI (such as DALL-E 3, ChatGPT, or a dedicated diagramming tool) to generate a high-end, professional architectural visualization.

Enhanced Prompt: RAG Architecture Diagram Generator

🎭 Role

Act as a Senior Solutions Architect and Technical Illustrator. Your specialty is creating clear, high-fidelity architectural diagrams for enterprise-level AI systems that are both aesthetically pleasing and technically accurate.

🌐 Context

We are documenting a robust Retrieval Augmented Generation (RAG) pipeline for a [PROJECT_NAME/STAKEHOLDER_GROUP]. The diagram needs to illustrate the data lifecycle—from ingestion and embedding to the sophisticated query-processing phase—to demonstrate technical maturity and system reliability.

🛠️ Task Instruction

Generate a high-quality, professional technical diagram illustrating a Retrieval Augmented Generation (RAG) architecture. Please organize the layout into four logical, flow-connected stages:

1. Ingestion & Embedding: Represent the flow from [SOURCE_TYPE] through Document Loader, Text Splitting, and Embedding Model.
2. Vector Persistence: Visualize the storage of embeddings into a Vector Database.
3. Advanced Retrieval: Illustrate the user query lifecycle including Query Rewriting/Expansion, Similarity Retrieval, and Re-ranking.
4. Generation: Depict the construction of the Contextual Prompt and the final output from the LLM.

⚖️ Constraints & Tone

• Style: Professional, clean, and modern. Avoid cluttered visuals or cartoonish elements.
• Color Palette: Use a cohesive professional palette based on [PRIMARY_COLOR] and [SECONDARY_COLOR] with smooth gradients (e.g., deep blue to vibrant violet).
• Icons: Use high-quality, minimalist, flat-style technical icons (e.g., vector database, document, AI brain, query icon).
• Clarity: Ensure clear directional arrows showing the data flow. Use white space effectively to avoid cognitive overload.
• Negative Constraints: Do not include unnecessary background noise or shadows. Do not include excessive text; keep labels concise and technical.

📝 Output Format

Provide an image (or a detailed Mermaid.js/PlantUML code block if requested) that represents the architecture. Ensure the composition is [ASPECT_RATIO, e.g., 16:9 for presentations or 1:1 for documentation].

🧩 Variables

• [PROJECT_NAME]: [Insert Name]
• [SOURCE_TYPE]: [e.g., PDF/Cloud Databases/API Streams]
• [PRIMARY_COLOR]: [e.g., Royal Blue]
• [SECONDARY_COLOR]: [e.g., Electric Violet]
• [ASPECT_RATIO]: [e.g., 16:9]

How to use this:

1. Copy the block above into your AI prompt window.
2. Fill in the [Variables] at the bottom before hitting enter.
3. For Image Generation (DALL-E 3): If the AI is not creating a perfect flowchart due to text rendering limitations, you may follow up with: "Great, now provide the same layout using clean labels and ensure the text is legible."
4. For Code-based Diagrams: If you prefer editable files, change the prompt to: "Provide this in Mermaid.js code syntax so I can render it in my documentation tool."