Navigating the AI Infrastructure Landscape: Lessons from Kubernetes Creators

2026-06-13 · Source: Stacklok · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure, Robotics & Autonomous Systems · Depth: Intermediate, long

Summary

Stacklok's CEO, Craig McLuckie and CTO, Joe Beda, shared insights on navigating the evolving enterprise AI infrastructure landscape, drawing parallels with the cloud-native era. They highlighted challenges like permissions, identity, and security, noting that vendors often push single-stack solutions while enterprises seek Kubernetes-like interoperability. The Model Context Protocol (MCP) is presented as a potential "Docker moment" for AI, offering a formalized protocol for controlled agent access to systems, critical for safety and productivity. The discussion also covered the shift in developer workflows, the need for guardrails to manage safety, effectiveness, and cost, and the importance of innovation "left of the LLM" for context and memory management. The authors advocate for open platforms to avoid vendor lock-in and emphasize starting with developer-centric paths to build trust before scaling AI agents to knowledge workers.

Key takeaway

For AI Architects and MLOps Engineers designing enterprise AI systems, prioritize open platform architectures and robust guardrails. Implement protocols like MCP for controlled agent access and deploy LLM gateways to manage costs and prevent data egress. This approach ensures interoperability and security, mitigating vendor lock-in risks while accelerating safe agent deployment for both developers and knowledge workers. Focus on building trust through controlled environments.

Key insights

The AI infrastructure landscape requires open platforms and guardrails like MCP to ensure interoperability, safety, and cost control, mirroring cloud-native lessons.

Principles

• Innovation often ignores convention.
• Guardrails enable velocity.
• Platform optionality beats lock-in.

Method

Start with developer-centric AI adoption to establish patterns and build trust. Implement coarse-grain controls and an LLM gateway for cost and security.

In practice

• Implement Model Context Protocol (MCP) for agent access control.
• Deploy an LLM gateway for cost control and data egress prevention.
• Start AI agent integration with developers.

Topics

• AI Infrastructure
• Model Context Protocol
• LLM Gateways
• Agentic Systems
• Cloud-Native Principles
• Enterprise AI
• Vendor Lock-in

Best for: CTO, VP of Engineering/Data, AI Product Manager, AI Architect, Director of AI/ML, MLOps Engineer

Related on AIssential

• Cloud Native Infrastructure Emerges as the Foundation for Trustworthy Agentic AI — Cloud-native infrastructure, utilizing Kubernetes and related tools, provides a robust foundation for trustworthy, production-grade agentic AI systems.
• Why Enterprises Move Beyond LiteLLM: The Case for an Enterprise MCP Platform in 2026 — Enterprises outgrow basic LLM proxies like LiteLLM when advanced security, cost attribution, agent governance, and Kubernetes-native operations become critical.
• MCP Dev Summit [Day 2] ft AWS, Docker, & Datadog — MCP is evolving to support real-time, interactive, and secure agent experiences, moving beyond simple request-response models.
• Agents think in milliseconds, legacy infrastructure doesn't. LinkedIn, Walmart and Zendesk shared how they closed the gap at VB Transform 2026 — Legacy enterprise infrastructure, built for human speeds, is the primary bottleneck for AI agents in production, not the models themselves.
• Designing AI Platforms That Scale: A Practical Blueprint — Centralize AI governance and observability to enable federated development and deployment, balancing control with speed.
• Persistent Systems, Kong join forces to secure enterprise AI scaling — Effective governance of intelligence flow across systems, not model access, defines enterprise AI success.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Stacklok.