Foundry IQ: RAG Without the Pain

The Problem: Every time you build an AI agent, it’s the same grind: connect data sources, write chunking logic, create embeddings, orchestrate retrieval pipelines. Rinse and repeat for every new agent.

The Nugget: Foundry IQ is Microsoft’s new intelligence layer for AI agents – powered by Azure AI Search. Instead of building custom RAG pipelines for each agent, you create a Knowledge Base once and connect as many agents as you need.

What Makes Foundry IQ Different:

Traditional RAGFoundry IQOne query → One index → DoneMulti-hop reasoning across sourcesCustom orchestration per agentOne Knowledge Base for all agentsManual chunking & vectorizingAutomatic indexing, vectorization, enrichment

Supported Knowledge Sources:

• Azure Blob Storage & OneLake

• SharePoint (M365)

• Azure AI Search Indexes

• Web (via Bing Grounding)

• MCP Servers (Preview)

Quick Setup in Foundry Portal:

• Open Microsoft Foundry Portal

• Create a Knowledge Base under your project

• Add Knowledge Sources (e.g., SharePoint site, Blob container)

• Set Retrieval Instructions in natural language

• Connect your agent via MCP Tool

Code Example (Python):

from agent_framework.azure import AzureAISearchContextProvider
from azure.identity.aio import DefaultAzureCredential

    endpoint="YOUR_SEARCH_ENDPOINT",
    knowledge_base_name="YOUR_KB_NAME",
    credential=DefaultAzureCredential(),
    mode="agentic"  *# Enable multi-hop reasoning*
) as search:
    *# Your agent can now access the Knowledge Base*

Why This Matters:

• +36% better answer quality according to Microsoft benchmarks

• Permission-aware – respects Purview labels & ACLs

• Single API endpoint for all your data sources

• Reusable – same Knowledge Base for multiple agents

Pro Tip: Use the retrieval_reasoning_effort setting: “minimal” for fast answers, “medium” for complex questions requiring multi-hop retrieval.

Admin context

For Microsoft admins, the practical point in Foundry IQ: RAG Without the Pain is to treat the change as something that should be validated before it becomes tenant-wide behavior. Check the affected users, devices, assignments and support process so the Nugget turns into a controlled operational improvement instead of another undocumented setting.

Runbook notes for Foundry IQ

Foundry IQ: RAG Without the Pain deserves a little more operational context because the decision usually affects AI knowledge retrieval. The related items are Foundry IQ, RAG, enterprise knowledge, permissions, grounding quality. Treat this Nugget as a starting point for a concrete tenant decision: who is in scope, which Microsoft portal or policy is touched, and what visible result should confirm that the configuration worked.

When validating Foundry IQ, keep the test narrow enough to understand the result. Select one representative user, device, workload or subscription, capture the current state, then apply the change and compare the outcome. This avoids guessing later when support sees a different enrollment state, access result, model response, update status or admin center signal.

The most useful documentation for Foundry IQ is practical rather than theoretical. Record the assignment logic, the owner, the expected monitoring view and the exception path. If the change affects users, include the wording support teams should use when they explain the behavior. If it affects devices or services, include the exact place where administrators can verify health.

For search consistency, keep the phrase Foundry IQ: RAG Without the Pain connected to the body text, the internal links and the category context. That helps readers understand why this Microsoft admin topic belongs with the surrounding Intune, Entra, Azure, Copilot, Security or automation Nuggets, and it gives AI search systems clearer signals about the real subject of the page.

Revisit Foundry IQ after the next rollout wave or Microsoft service update. Cloud behavior, licensing boundaries and portal labels can move quickly, so a short review prevents stale instructions. Confirm that the original assumption is still true, remove obsolete exceptions, and update the runbook if the operating model changed.

A clean handover for Foundry IQ should also include a fallback. Write down how the team pauses the change, narrows the scope, or returns to the previous configuration if the result creates noise. This makes the Nugget safer to use in production because the implementation path includes both the happy path and the recovery path.