Skip to main content
Practical AI

Rebooting Enterprise AI with MCP and Kubernetes

48 min episode · 2 min read
·
Craig Mcluckie

Episode

48 min

Read time

2 min

Topics

Productivity, Startups, Leadership

AI-Generated Summary

Key Takeaways

  • MCP Platform Stack: Enterprises need four components to operationalize MCP: a secure runtime (containerized via OCI images), a vetted registry of approved servers, a gateway providing a single endpoint across models like Claude and GPT, and a control plane for mapping servers to user groups. Building these once enables multi-model access without vendor lock-in.
  • Tool Pollution Reduction: When agents access three to four MCP servers, context windows can carry 150+ tools consuming 20,000–30,000 tokens per interaction. Replacing direct tool exposure with two proxy endpoints — find-tool and bulk-tool — reduces input token consumption by 80–90%, dramatically improving smaller model reliability from unpredictable rates to 95–97% accuracy.
  • Identity and Token Exchange: Rather than passing raw user credentials to MCP servers, enterprises should implement token exchange at the proxy layer — mapping Okta or OIDC tokens to descoped API keys or federated credentials. This pattern enforces least-privilege access, such as read-only AWS permissions, without requiring individual MCP server developers to handle authentication mechanics.
  • Agentic Concurrency as Productivity Multiplier: Stacklock's engineering team runs 5–15 simultaneous agents, each with distinct configuration and tool access, managed by one human operator. This approach produced a 60% throughput increase in a single week, enabling the team to resolve incoming issues faster than they accumulate — a pattern Craig Mc expects to transfer to non-developer knowledge workers.
  • Kubernetes for MCP Orchestration: ToolHive, an Apache 2.0 open-source project, containerizes MCP servers using OCI images, enabling file system and network endpoint restrictions per server. Kubernetes adoption for running MCP servers is growing 50% month-over-month, with millions of tool invocations now originating from Kubernetes deployments rather than developer desktops.

What It Covers

Craig Mc, CEO of Stacklock, explains how Model Context Protocol (MCP) functions as enterprise infrastructure for agentic AI systems, covering the four-component platform stack — runtime, registry, gateway, and control plane — and how Kubernetes provides the orchestration layer for scaling MCP servers across organizations.

Key Questions Answered

  • MCP Platform Stack: Enterprises need four components to operationalize MCP: a secure runtime (containerized via OCI images), a vetted registry of approved servers, a gateway providing a single endpoint across models like Claude and GPT, and a control plane for mapping servers to user groups. Building these once enables multi-model access without vendor lock-in.
  • Tool Pollution Reduction: When agents access three to four MCP servers, context windows can carry 150+ tools consuming 20,000–30,000 tokens per interaction. Replacing direct tool exposure with two proxy endpoints — find-tool and bulk-tool — reduces input token consumption by 80–90%, dramatically improving smaller model reliability from unpredictable rates to 95–97% accuracy.
  • Identity and Token Exchange: Rather than passing raw user credentials to MCP servers, enterprises should implement token exchange at the proxy layer — mapping Okta or OIDC tokens to descoped API keys or federated credentials. This pattern enforces least-privilege access, such as read-only AWS permissions, without requiring individual MCP server developers to handle authentication mechanics.
  • Agentic Concurrency as Productivity Multiplier: Stacklock's engineering team runs 5–15 simultaneous agents, each with distinct configuration and tool access, managed by one human operator. This approach produced a 60% throughput increase in a single week, enabling the team to resolve incoming issues faster than they accumulate — a pattern Craig Mc expects to transfer to non-developer knowledge workers.
  • Kubernetes for MCP Orchestration: ToolHive, an Apache 2.0 open-source project, containerizes MCP servers using OCI images, enabling file system and network endpoint restrictions per server. Kubernetes adoption for running MCP servers is growing 50% month-over-month, with millions of tool invocations now originating from Kubernetes deployments rather than developer desktops.

Notable Moment

Craig Mc describes watching a non-technical go-to-market colleague approach an engineer to configure five MCP servers — HubSpot, LinkedIn, and others — for their own workflow, signaling that MCP adoption has moved well beyond developer tooling into general business operations.

Know someone who'd find this useful?

Episode Transcript

Welcome to the Practical AI podcast, where we break down the real world applications of artificial intelligence and how it's shaping the way we live, work, and create. Our goal is to help make AI technology practical, productive, and accessible to everyone. Whether you're a developer, business leader, or just curious about the tech behind the buzz, you're in the right place. Be sure to connect with us on LinkedIn, X, or Blue Sky to stay up to date with episode drops, behind the scenes content, and AI insights. You can learn more at practicalai.fn. Now on to the show. Welcome to another episode of the Practical AI podcast. This is Daniel Leitnak. I am CEO of Prediction Guard, and I'm joined as always by my co host, Chris Benson, who is a principal AI and autonomy research engineer. How are you doing, Chris? Doing great today, Daniel. How's it going? It's it's going it's going really good. I I feel like I'm going to, you know, just add a bunch of tools to my to my toolkit today, via via MCP, because today, we have with us Craig McLuckie, who is CEO of Stack Lock, which we'll learn a little bit more about and, of course, talk a a good bit about MCP and other things. Welcome, Craig. It's great to have you. Hey. Thanks for having me on the show. Yeah. Well, I I kind of I mean, and people can look, look up the the projects, that you're working on now in Stack Lock. A lot of that has to do with MCP and AI on top of Kubernetes. Do you wanna just give us a a little bit of a of a kind of setup of why you are spending your time thinking about this intersection of AI, MCP, Kubernetes? Yeah. I mean, I think for me, it's it's sort of interesting. And my you know, maybe I can just introduce my background a little bit. You know, I'm an infrastructure guy. I've been building infrastructure technology for pretty much the totality of my career. You know, I built a lot of infrastructure tech when I was at Microsoft. And then at Google, I was, happy to meet my friend Joe, and we built out a Google Compute Engine, which is, you know, virtual machine infrastructure technology. We started the Kubernetes project, did a few other things, you know, inside there. We and subsequently, we've built a lot of other, technologies together, you know, either as in the startup context or or as part of, larger companies. You know, I think for me, there's this sort of a lot of interesting parallels. I think, you know, history doesn't repeat itself, but it often rhymes. And, you know, one of the things that, you know, that got me really kinda captivated by this whole MTP thing was you know, taking me back to my earlier career, I remember the first time I saw Docker. And so Docker …

Get the full transcript (10,220 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Practical AI transcripts →

You just read a 3-minute summary of a 45-minute episode.

Get Practical AI summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links.

Tools

  • Craig Mc, CEO of Stacklock, explains how Model Context Protocol (MCP) functions as enterprise infrastructure for agentic AI systems
  • Kubernetes adoption for running MCP servers is growing 50% month-over-month, with millions of tool invocations now originating from Kubernetes deployments
  • by Anthropic

    a gateway providing a single endpoint across models like Claude and GPT
  • Craig Mc describes watching a non-technical go-to-market colleague approach an engineer to configure five MCP servers — HubSpot, LinkedIn, and others
  • SPONSORS: Prediction Guard
  • ToolHive, an Apache 2.0 open-source project, containerizes MCP servers using OCI images, enabling file system and network endpoint restrictions per server.
  • Rather than passing raw user credentials to MCP servers, enterprises should implement token exchange at the proxy layer — mapping Okta or OIDC tokens to descoped API keys
  • by OpenAI

    a gateway providing a single endpoint across models like Claude and GPT

company

  • Craig Mc, CEO of Stacklock, explains how Model Context Protocol (MCP) functions as enterprise infrastructure

More from Practical AI

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best AI Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's Startups & Product Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into Practical AI.

Every Monday, we deliver AI summaries of the latest episodes from Practical AI and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime