{"slug": "ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag", "title": "Ask HN: Should I Combine Global Knowledge, Internet Search, and User RAG", "summary": "A developer building a SaaS platform in Sri Lanka for handling sensitive documents is seeking advice on whether to combine global knowledge, internet search, and user-specific RAG. The developer is considering two architectures: Option 1 uses a base LLM with a platform-managed global knowledge base and user-specific RAG, while Option 2 involves fine-tuning an open-source LLM on domain-specific data plus user RAG. The developer leans toward Option 1 due to concerns about the cost and complexity of fine-tuning.", "body_md": "I'm building a SaaS platform in Sri Lanka that handles documents and other sensitive data.\n\nEach user can upload their own documents and information, and the platform uses RAG to answer questions based on that user's data. That part makes sense to me.\n\nMy main concern is what happens when the user hasn't uploaded enough information. I still want the LLM to provide accurate answers using reliable information from the internet (or from a curated knowledge base), with proper citations.\n\nThese are the two architectures I'm considering:\n\nOption 1:\n\nBase LLM (OpenAI/Anthropic via Azure AI Foundry or Amazon Bedrock) ↓ Platform RAG (global knowledge base managed by us) ↓ User-specific RAG In this approach, we maintain a global knowledge base that we (the platform admins) curate and update. Every user can access this shared knowledge, while their own uploaded documents are searched through their personal RAG.\n\nOption 2:\n\nOpen-source LLM ↓ Fine-tuned on Sri Lankan/domain-specific data ↓ User-specific RAG Here, we fine-tune an open-source model using Sri Lankan or domain-specific data, and each user still has their own RAG for their private documents.\n\nMy concerns are:\n\nIs fine-tuning actually the right solution here, or is it unnecessary?\n\nIs a global/shared RAG a better approach than fine-tuning?\n\nHow would you design this architecture if you wanted:\n\nAccurate answers from domain knowledge\n\nUser-private document search\n\nCitations/sources\n\nGood scalability for thousands of users\n\nI'm leaning toward Option 1 because fine-tuning seems expensive, time-consuming, and I have no experience with it yet. However, I'm not sure if I'm thinking about this correctly.\n\nI'd really appreciate hearing how others would approach this problem\n\nComments URL: [https://news.ycombinator.com/item?id=49059969](https://news.ycombinator.com/item?id=49059969)\n\nPoints: 1\n\n# Comments: 1", "url": "https://wpnews.pro/news/ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag", "canonical_source": "https://news.ycombinator.com/item?id=49059969", "published_at": "2026-07-26 16:55:11+00:00", "updated_at": "2026-07-26 17:22:21.691108+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-startups", "ai-products"], "entities": ["OpenAI", "Anthropic", "Azure AI Foundry", "Amazon Bedrock"], "alternates": {"html": "https://wpnews.pro/news/ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag", "markdown": "https://wpnews.pro/news/ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag.md", "text": "https://wpnews.pro/news/ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag.txt", "jsonld": "https://wpnews.pro/news/ask-hn-should-i-combine-global-knowledge-internet-search-and-user-rag.jsonld"}}