AI agents are rapidly becoming a new way to build applications. But for agents to be truly useful, they need access to the knowledge and context that helps them reason about the world they operate in.

That’s where Foundry IQ comes in.

Today we’re announcing the IQ Series: Foundry IQ, a new set of developer-focused episodes exploring how to build knowledge-centric AI systems using Foundry IQ.

The series focuses on the core ideas behind how modern AI systems work with knowledge, how they retrieve information, reason across sources, synthesize answers, and orchestrate multi-step interactions.

Instead of treating retrieval as a single step in a pipeline, Foundry IQ approaches knowledge as something that AI systems actively work with throughout the reasoning process. The IQ Series breaks down these concepts and shows how they come together when building real AI applications.

You can explore the series and all the accompanying samples here:

👉 https://aka.ms/iq-series

What is Foundry IQ?

Foundry IQ helps AI systems work with knowledge in a more structured and intentional way.

Rather than wiring retrieval logic directly into every application, developers can define knowledge bases that connect to documents, data sources, and other information systems. AI agents can then query these knowledge bases to gather the context they need to generate responses, make decisions, or complete tasks.

This model allows knowledge to be organized, reused, and combined across applications, instead of being rebuilt for each new scenario.

What's covered in the IQ Series?

The Foundry IQ episodes in the IQ Series explore the key building blocks behind knowledge-driven AI systems from how knowledge enters the system to how agents ultimately query and use it.

The series is released as three weekly episodes:

• Foundry IQ: Unlocking Knowledge for Your Agents — March 18, 2026: Introduces Foundry IQ and the core ideas behind it. The episode explains how AI agents work with knowledge and walks through the main components of the Foundry IQ that support knowledge-driven applications.

• Foundry IQ: Building the Data Pipeline with Knowledge Sources — March 25, 2026: Focuses on Knowledge Sources and how different types of content flow into Foundry IQ. It explores how systems such as SharePoint, Fabric, OneLake, Azure Blob Storage, Azure AI Search, and the web contribute information that AI systems can later retrieve and use.

• Foundry IQ: Querying the Multi-Source AI Knowledge Bases — April 1, 2026: Dives into the Knowledge Bases and how multiple knowledge sources can be organized behind a single endpoint. The episode demonstrates how AI systems query across these sources and synthesize information to answer complex questions. 

Each episode includes a short executive introduction, a tech talk exploring the topic in depth, and a visual recap with doodle summaries of the key ideas.

Alongside the episodes, the GitHub repository provides cookbooks with sample code, summary of the episodes, and additinal learning resources, so developers can explore the concepts and apply them in their own projects.

Explore the Repo

All episodes and supporting materials live in the IQ Series repository:

👉 https://aka.ms/iq-series

Inside the repository you’ll find:

• The Foundry IQ episode links
• Cookbooks for each episode
• Links to documentation and additional resources

If you're building AI agents or exploring how AI systems can work with knowledge, the IQ Series is a great place to start.

Watch the episodes and explore the cookbooks! We’re excited to see what you build and welcome your feedback & ideas as the series evolves.

If you’re learning to build AI agents, you’ve probably hit a familiar wall: your agent can generate text, but it doesn’t actually know anything about your data. It can’t look up your documents, search across your files, or pull facts from multiple sources to answer a real question.

That’s the gap Foundry IQ fills. It gives your AI agents structured access to knowledge, so they can retrieve, reason over, and synthesize information from real data sources instead of relying on what’s baked into the model.

Why Should You Care?

As a student or early-career developer, understanding how AI systems work with external knowledge is one of the most valuable skills you can build right now. Retrieval-Augmented Generation (RAG), knowledge bases, and multi-source querying are at the core of every production AI application, from customer support bots to research assistants to enterprise copilots.

Foundry IQ gives you a hands-on way to learn these patterns without having to build all the plumbing yourself. You define knowledge bases, connect data sources, and let your agents query them. The concepts you learn here transfer directly to real-world AI engineering roles.

What is Foundry IQ?

Foundry IQ is a service within Azure AI Foundry that lets you create knowledge bases, collections of connected data sources that your AI agents can query through a single endpoint.

Instead of writing custom retrieval logic for every app you build, you:

1. Define knowledge sources — connect documents, data stores, or web content (SharePoint, Azure Blob Storage, Azure AI Search, Fabric OneLake, and more).
2. Organize them into a knowledge base — group multiple sources behind one queryable endpoint.
3. Query from your agent — your AI agent calls the knowledge base to get the context it needs before generating a response.

This approach means the knowledge layer is reusable. Build it once, and any agent or app in your project can tap into it.

The IQ Series: A Three-Part Learning Path

The IQ Series is a set of three weekly episodes that walk you through Foundry IQ from concept to code. Each episode includes a tech talk, visual doodle summaries, and a companion cookbook with sample code you can run yourself.

👉 Get started: https://aka.ms/iq-series

Episode 1: Unlocking Knowledge for Your Agents (March 18, 2026)

Start here. This episode introduces the core architecture of Foundry IQ and explains how AI agents interact with knowledge. You’ll learn what knowledge bases are, why they matter, and how the key components fit together.

What you’ll learn:

• The difference between model knowledge and retrieved knowledge
• How Foundry IQ structures the retrieval layer
• The building blocks: knowledge sources, knowledge bases, and agent queries

Episode 2: Building the Data Pipeline with Knowledge Sources (March 25, 2026)

This episode goes deeper into knowledge sources, the connectors that bring data into Foundry IQ. You’ll see how different content types flow into the system and how to wire up sources from services you may already be using.

What you’ll learn:

• How to connect sources like Azure Blob Storage, Azure AI Search, SharePoint, Fabric OneLake, and the web
• How content is ingested and indexed for retrieval
• Patterns for combining multiple source types

Episode 3: Querying Multi-Source Knowledge Bases (April 1, 2026)

The final episode shows you how to bring it all together. You’ll learn how agents query across multiple knowledge sources through a single knowledge base endpoint and how to synthesize answers from diverse data.

What you’ll learn:

• How to query a knowledge base from your agent code
• How retrieval works across multiple connected sources
• Techniques for synthesizing information to answer complex questions

Get Hands-On with the Cookbooks

Every episode comes with a companion cookbook in the GitHub repo, complete with sample code you can clone, run, and modify. This is the fastest way to go from watching to building.

👉 Explore the repo: https://aka.ms/iq-series

Inside you’ll find:

• Episode links — watch the tech talks and doodle recaps
• Cookbooks — step-by-step code samples for each episode
• Documentation links — official Foundry IQ docs and additional learning resources

What to Build Next

Once you’ve worked through the series, try applying what you’ve learned:

• Study assistant — connect your course materials as knowledge sources and build an agent that can answer questions across all your notes and readings.
• Project documentation bot — index your team’s project docs and READMEs into a knowledge base so everyone can query them naturally.
• Research synthesizer — connect multiple data sources (papers, web content, datasets) and build an agent that can cross-reference and summarize findings.

Start Learning

The IQ Series is designed to take you from zero to building knowledge-driven AI agents. Watch the episodes, run the cookbooks, and start experimenting with your own knowledge bases.

👉 https://aka.ms/iq-series

1168. This week, we look at the word "leprechaun" and its surprisingly wild origin story involving shoemaking, ancient Rome, and wolf-men. Then we look at the word "equinox": its Chaucer connection, the newer word "equilux," and why the first point of Aries is actually in Pisces now (and headed for Aquarius).

🔗 Join the Grammar Girl Patreon.

🔗 Share your familect recording in Speakpipe or by leaving a voicemail at 833-214-GIRL (833-214-4475)

🔗 Watch my LinkedIn Learning writing courses.

🔗 Subscribe to the newsletter.

🔗 Take our advertising survey. 

🔗 Get the edited transcript.

🔗 Get Grammar Girl books. 

| HOST: Mignon Fogarty

| Grammar Girl is part of the Quick and Dirty Tips podcast network.

• Audio Engineer: Castria Communications
• Director of Podcast: Holly Hutchings
• Advertising Operations Specialist: Morgan Christianson
• Marketing and Video: Nat Hoopes, Rebekah Sebastian
• Podcast Associate: Maram Elnagheeb

| Theme music by Catherine Rannus.

| Grammar Girl Social Media: YouTube. TikTok. Facebook. Threads. Instagram. LinkedIn. Mastodon. Bluesky.

Download audio: https://dts.podtrac.com/redirect.mp3/media.blubrry.com/grammargirl/cdn.simplecast.com/media/audio/transcoded/dd74e7bd-f654-43a6-b249-3f071c897900/e7b2fc84-d82d-4b4d-980c-6414facd80c3/episodes/audio/group/acb40d66-bfac-4dc9-9a76-d09ab56a8cd0/group-item/e2736721-7a98-4801-b2d3-1c3b906e0162/128_default_tc.mp3?aid=rss_feed&feed=XcH2p3Ah

In this episode, José Quaresma speaks with Iris Dyrmishi, Senior Observability Engineer at Miro, about building an observability platform that hundreds of engineers actually trust and use. Iris explains how her team treats observability as an internal product, walks through Miro's tracing migration from Jaeger and Zipkin to OpenTelemetry with zero disruption, and shares how teams now use traces proactively to find bottlenecks before they become outages. The conversation also covers the honest downsides — alert noise, dashboard sprawl, and the cost of observability — including a recent example using eBPF and Grafana Beyla to uncover hidden networking expenses that transformed Miro's cloud bill.

Episode page

---

• (00:00) - Intro
• (00:59) - Building Observability as a Product at Miro
• (04:08) - Migrating to OpenTelemetry
• (09:21) - Industry Maturity and the Business Case
• (12:02) - From Reactive to Proactive Observability
• (14:34) - Logs vs. Tracing Explained
• (18:04) - Team Ownership, AI, and Freedom
• (24:38) - The Downsides and Costs of Observability
• (29:58) - Rapid Fire and Close

Iris Dyrmishi is a Senior Observability Engineer at Miro, where she builds and maintains the company's observability platform. She started as a backend engineer before moving into SRE roles at Worten Portugal and Farfetch, where she developed her specialty in tracing and drove OpenTelemetry migrations across large engineering organisations without disrupting existing workflows. A CNCF Ambassador, co-organiser of Kubernetes Community Days Porto, and active voice in the observability community, she writes extensively about practical adoption challenges and has spoken at KubeCon EU and on the o11ycast podcast. Her guiding philosophy: observability is a team sport.

This podcast is hosted by José Quaresma, researched by Joseph Thwaites and produced by Perseu Mandillo. 

© Queue-it, 2026

Download audio: https://media.transistor.fm/c795f114/04dc92f5.mp3

The hard part of AI coding isn't generating code — it's controlling quality, safety, and drift. Kaushik and Iury break down harness engineering: the five pillars for shaping an agent's environment and what it looks like when teams build custom harnesses from scratch.

Full shownotes at fragmentedpodcast.com.

Show Notes

Why it matters

• Harness Engineering -
  OpenAI's post on building their Codex codebase (~1M lines of code, 1,500 PRs
  merged, zero manually written)

Shaping the harness

• The Feed's Lost and Found -
  Iury's newsletter consolidating harness engineering themes

1. Agent legibility
2. Closed feedback loops
3. Persistent memory
4. Entropy control
5. Blast radius controls

Building the harness

• Minions: Stripe's one-shot, end-to-end coding agents -
  Stripe forked Goose to build custom agents for their codebase
• Goose - open-source coding agent from Block
• Superpowers by Jesse Vincent - skills
  that enforce a proper software engineering process
• Open Code - open-source coding agent you can fork and
  customize

Other resources

• Agent Harness Glossary -
  Latent Patterns
• Towards self-driving codebases -
  Cursor
• Agentic Workflows -
  GitHub Next
• Future of Software Development -
  ThoughtWorks

Get in touch

We'd love to hear from you. Email is the
best way to reach us or you can check our contact page for other
ways.

We want to hear all the feedback: what's working, what's not, topics you'd like
to hear more on.

• Contact us
• Newsletter
• Youtube
• Website

Co-hosts:

• Kaushik Gopal
• Iury Souza

[!fyi] We transitioned from Android development to AI starting with
Ep. #300. Listen to that episode for the full story behind
our new direction.

Download audio: https://cdn.simplecast.com/media/audio/transcoded/eea5748e-09ed-4c46-8f6a-b2569dd75851/20f35050-e836-44cd-8f7f-fd13e8cb2e44/episodes/audio/group/face4e8b-6fb2-4f91-ab8e-8ad35b829715/group-item/0ce6c86f-d6ef-4ee1-8fa7-9ac6958feed6/128_default_tc.mp3?aid=rss_feed&feed=LpAGSLnY

Show Notes – Episode #242 In the world of business communication, the line between a presentation and a document is often blurred—especially when PowerPoint is the authoring tool of choice. In episode 242, our three industry experts—Troy Chollar (TLC Creative Services), Sandy Johnson (Presentation Wiz), and Nolan Haims (Nolan Haims Creative) dive [...]



Download audio: https://traffic.libsyn.com/thepresentationpodcast/TPP_e242.mp3