AI Agents · Lesson

Combining Web Search with RAG

Hybrid retrieval: local vector store + live web search for up-to-date answers.

The Hybrid Retrieval Challenge

Most real-world agents need two types of knowledge: static domain knowledge (company docs, product manuals, policies) and current information (news, live prices, recent events).

A local vector store handles the former; web search handles the latter. Combining both gives the best of both worlds.

Architecture: Local RAG + Web Search

The hybrid system routes each question to the right retrieval source:

Vector store (RAG) — indexed documents, stable knowledge, private data
Web search — current events, recent releases, live data
Both — when the question needs context from docs AND current info

class HybridRetrievalAgent:
    def __init__(self, vector_store, search_client):
        self.vector_store = vector_store  # e.g., ChromaDB or FAISS
        self.search_client = search_client  # e.g., TavilyClient

    def answer(self, question):
        route = self.classify_question(question)

        if route == 'static':
            context = self.vector_store.query(question, n_results=5)
        elif route == 'current':
            context = self.web_search(question)
        else:  # 'both'
            local = self.vector_store.query(question, n_results=3)
            web = self.web_search(question)
            context = local + web

        return self.generate_answer(question, context)

All lessons in this course

← Back to AI Agents