Building the Foundation for the Agentic AI Era
Angie Jones, VP of the Agentic AI Foundation, talks with Chris about open standards and projects such as MCP, A2A and Goose, and about getting agents, tools and systems to work together at scale.
Listening guide
Podcasts like Practical AI
Practical AI is Daniel Whitenack and Chris Benson's show about making AI practical, with guest interviews and episodes where the two hosts talk a topic through. Here are six of its episodes, each with a Chain of Thought episode to play next.
- 6 episode pairings
I host Chain of Thought. We have no affiliation with Practical AI: no partnership, no arrangement, nothing either of us gets from this page.
What Practical AI is
Practical AI brings technology professionals, business people, students, enthusiasts and expert guests together to discuss AI and related topics. Its stated focus is productive implementations and real-world scenarios accessible to everyone.
Daniel Whitenack and Chris Benson host guest interviews and host-only episodes, including their Fully Connected conversations. Recent subjects include enterprise agents, agent security, open models, physical AI, image generation, AI policy and work. The first episode came out in July 2018, and episode 375 came out on 8 October 2026. Independent now, Practical AI used to be part of the Changelog podcast family.
At a glance
| Detail | Practical AI | Chain of Thought |
|---|---|---|
| Hosts | Daniel Whitenack and Chris Benson | Conor Bronsdon |
| First episode | July 2018 | November 2024 |
| Format | Guest interviews, plus host-only 'Fully Connected' episodes | Interviews with engineers, founders and operators shipping AI in production |
| Typical length | 47 minutes (median, April to October 2026) | 53 minutes (median, April to October 2026) |
| Transcripts | On episode pages for recent episodes | A full transcript on every episode page |
| Where to listen | practicalai.show | Episode archive and podcast apps |
Figures for Practical AI come from its own site and feed, checked 14 October 2026.
Where Chain of Thought fits
Chain of Thought is a podcast about AI infrastructure and the people building it. Conor Bronsdon interviews engineers, founders and operators shipping AI in production, and each conversation tends to stay on one company's decisions: what it built, what it cost and what it would change. Practical AI has two hosts and mixes interviews with host-only episodes; Chain of Thought has one host and focuses on interviews.
Which one to start with
Start with Practical AI if you want two hosts covering a wide range of AI as it moves, from agents and open models to physical AI and policy, in guest interviews and in host-only episodes such as their walk-through of Anthropic's Zero Trust for AI Agents framework.
Start with Chain of Thought if you want to hear from the team that shipped the thing: what they built, what it cost and what they would change.
You do not have to pick one. The pairings below are where the two shows meet.
-
Then play
How Block Deployed AI Agents to 12,000 Employees in 8 Weeks w/ MCP | Angie Jones Angie Jones, Block Angie Jones was on Chain of Thought in January 2026, explaining how her team at Block used MCP and Goose to put AI agents in front of 12,000 employees in eight weeks. -
Computer-Use Agents and the Future of the Agentic Internet
Demetrios Brinkmann discusses computer-use agents, MCP, agent harnesses and agent-to-agent interactions, including the hurdles to enterprise use.
Then play
Most of the Web Will Never Get APIs for AI Agents | Dhruv Batra, Yutori Dhruv Batra, Yutori Dhruv Batra of Yutori takes the computer-use question to the web itself, where his browser agents read pixels and click buttons on sites without APIs. -
Zero Trust for AI Agents
Daniel and Chris work through Anthropic's Zero Trust for AI Agents framework, the risks it addresses and controls for agentic systems.
Then play
We Built Agents, Nobody Built HR | Tyler Akidau, Redpanda Tyler Akidau, Redpanda Tyler Akidau of Redpanda turns those security concerns into four governance jobs for deployed agents: identity, authorization, observability and accountability. -
Less about Models; More about Architecture
Rackspace Chief AI Officer Chetan Gupta discusses architecture, model deployment, governance and AI sovereignty as organizations move AI from experiments into enterprise deployment.
Then play
Architecting AI Agents: The Shift from Models to Systems | Aishwarya Srinivasan Aishwarya Srinivasan, Fireworks AI Aishwarya Srinivasan of Fireworks AI follows the architecture question into production agents built from model orchestration, memory, tool calls and evaluation. -
The Myth of Model Wars: Open vs Closed AI in 2026
Daniel and Chris ask whether the open versus closed model split still matters, from physical AI and edge devices to open models such as Llama, then turn to agentic systems and workflows.
Then play
How Intercom Cut $250K/Month by Ditching GPT for Open Models | Fergal Reid Fergal Reid, Intercom Fergal Reid of Intercom puts open model choice into a production budget, explaining why his team moved a summarization task he put at about $250K a month from GPT to a post-trained 14B Qwen model. -
AI Proficiency: From Users to Builders
Mike Lewis of TiER1 Performance walks through an L0 to L3 framework for workforce AI proficiency, with a focus on non-technical builders, resistance to AI and measurable business value.
Then play
Stop Token Maxxing: Find Where AI Actually Pays Off | Jiaona Zhang, Laurel Jiaona Zhang, Laurel Jiaona Zhang of Laurel asks teams to measure where AI returns time and explains how people in roles such as customer success and design ship features.
Sources
Facts about Practical AI on this page were checked against these pages on 14 October 2026.