Forget CLIs and GUIs: AI is the New Interface for Developer Platforms

Discover how AI can revolutionize user interactions with Internal Developer Platforms. This video demonstrates using AI to create, observe, and delete services through natural language conversations. Watch as we leverage custom AI commands to dynamically generate Kubernetes manifests, monitor service status, and manage resources, all without users needing deep platform knowledge. Learn about the pros and cons of this approach, including flexibility vs. determinism. See how AI can simplify complex operations, making platform interactions more intuitive for developers. This approach showcases the potential of AI in streamlining DevOps processes and enhancing developer experiences. We'll use Claude Code custom commands today, but the same principles could be applied to other agents as well.

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Sponsor: TestSprite 🔗 https://testsprite.com ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

InternalDeveloperPlatform, #AI, #PlatformEngineering

Consider joining the channel: https://www.youtube.com/c/devopstoolkit/join

▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/internal-developer-platforms/forget-clis-and-guis-ai-is-the-new-interface-for-developer-platforms 🔗 Claude Code: https://claude.ai 🎬 Claude Code: AI Agent for DevOps, SRE, and Platform Engineering: https://youtu.be/h-6LP133o6w

▬▬▬▬▬▬ 💰 Sponsorships 💰 ▬▬▬▬▬▬ If you are interested in sponsoring this channel, please visit https://devopstoolkit.live/sponsor for more information. Alternatively, feel free to contact me over Twitter or LinkedIn (see below).

▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social ➡ LinkedIn: https://www.linkedin.com/in/viktorfarcic/

▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬ 🎤 Podcast: https://www.devopsparadox.com/ 💬 Live streams: https://www.youtube.com/c/DevOpsParadox

▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬ 00:00 AI for Internal Developer Platforms (IDPs) 01:40 TestSprite (sponsor) 02:41 AI for Internal Developer Platforms (IDPs) (cont.) 03:25 AI to Create Platform Services 08:21 AI to Observe Platform Services 10:32 AI to Fix Issues with Platform Services 11:31 AI to Fix Delete Platform Services 12:52 How Did It All Happen? 14:43 Pros and Cons of AI as Platform Interface

via YouTube https://www.youtube.com/watch?v=ApjnCa-a2xI

Ep23 - Ask Me Anything About Anything with Esmira Bayramova

There are no restrictions in this AMA session. You can ask anything about DevOps, Cloud, Kubernetes, Platform Engineering, containers, or anything else. We'll have special guest Esmira Bayramova to help us out.

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Sponsor: Codefresh 🔗 GitOps Argo CD Certifications: https://learning.codefresh.io (use "viktor" for a 50% discount) ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social ➡ LinkedIn: https://www.linkedin.com/in/viktorfarcic/

▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬ 🎤 Podcast: https://www.devopsparadox.com/ 💬 Live streams: https://www.youtube.com/c/DevOpsParadox

via YouTube https://www.youtube.com/watch?v=b5pN35kcOkk

Week Ending May 25, 2025

https://lwkd.info/2025/20250527

Developer News

The Program Committee is now accepting applications for the Maintainer Summit North America 2025. Share your interest in joining the committee before Monday, July 7th.

Release Schedule

Next Deadline: PRR Freeze, June 12th

The Release Cycle for 1.34 has started, and the release team is actively collecting enhancements. SIG Leads should discuss enhancements and add the lead-opted-in label for KEPs going into v1.34.

Featured PRs

131842: Add metrics for compatibility version

This PR adds alpha metrics for binary, emulation, and minimum compatibility versions in componentGlobalsRegistry, exposed via Prometheus in kube-apiserver, scheduler, and controller-manager for observability of version negotiation. It introduces an AddMetrics method that publishes the binary version, emulation version, and minimum compatibility version of each component as Prometheus gauge metrics. Users can now monitor version negotiation for kube-apiserver, scheduler, and controller-manager using these metrics.

128748: feat: introduce pInfo.UnschedulableCount to make the backoff calculation more appropriate

This PR updates the scheduler to separate scheduling failures caused by plugin rejections from those caused by internal errors. It introduces UnschedulableCount to track only plugin-based rejections, ensuring that transient errors like API failures or network issues do not increase backoff time unfairly. This change improves scheduling fairness and responsiveness under cluster instability.

129983: feature(scheduler): Customizable pod selection and ordering in DefaultPreemption plugin

This PR introduces support for customizing pod selection and ordering in the DefaultPreemption plugin; It adds optional EligiblePods and OrderedPods function hooks, allowing scheduler integrations to override the default behavior without reimplementing the plugin. This enables more flexible preemption strategies while maintaining the existing plugin interface.

This PR adds support for the EncryptionAlgorithmECDSAP384 in kubeadm API types; Users can now choose ECDSA-P384 for generating PKI assets like CA and component certificates during kubeadm init; Implemented key generation logic for ECDSA P-384 keys in pkiutil (using elliptic.P384()). This ensures the algorithm is handled correctly across pkiutil and cluster configuration paths.

KEP of the Week

KEP 4369: Allow almost all printable ASCII characters in environment variables

This enhancement allowed all printable ASCII characters (with ASCII codes 32–126), except "=", to be used in environment variable names. Previously, Kubernetes imposed restrictions that could prevent certain applications from functioning as intended, especially when users couldn’t control the variable names. By lifting these constraints, the change improved compatibility with a broader range of applications and removed an adoption barrier, aligning Kubernetes behaviour more closely with real-world usage patterns

This KEP is tracked for beta in v1.34.

Other Merges

automatic_reloads of authz config metrics to beta

Pod backoff to be completely skipped when PodMaxBackoffDuration kube-scheduler option is set to zero

Shorthand for –output flag in kubectl explain which was accidentally deleted has been added back

Kubernetes is now built using Go 1.24.3

References to group resource in metrics unified

e2e: Shadowed error fixed in reboot test

Filter integration tests added for NodeAffinity plugin

AuthenticationConfiguration type has been promoted to apiserver.config.k8s.io/v1

Volumes on nodes to not be expanded if controller expansion is finished

Promotions

QueueingHint to GA

kuberc to beta

Version Updates

system-validators to v1.10.1

etcd to v3.6.0

Go for publishing bot rules to 1.23.9

Subprojects and Dependency Updates

minikube v1.36.0 delivers significantly faster vfkit networking on macOS with the --network vmnet-shared option, supports Kubernetes v1.33.1, enables addon configuration via a dedicated config file, and includes additional improvements

vertical-pod-autoscaler v1.4.0 is out, with alpha support for in-place pod resource updates via the InPlaceOrRecreate Feature Gate, improved resource tracking from pod status, options for global maximum resource limits, and a set of bug fixes and dependency updates

kubespray v2.28.0 is out with a bunch of version updates. Krew installation support is removed.

Shoutouts

No shoutouts this week. Want to thank someone for special efforts to improve Kubernetes? Tag them in the #shoutouts channel.

via Last Week in Kubernetes Development https://lwkd.info/

May 27, 2025 at 06:10PM

Performance testing Kubernetes workloads, with Stephan Schwarz

https://ku.bz/yY-FnmGfH

If you're tasked with performance testing Kubernetes workloads without much guidance, this episode offers clear, experience-based strategies that go beyond theory.

Stephan Schwarz, a DevOps engineer at iits-consulting, walks through his systematic approach to performance testing Kubernetes applications. He covers everything from defining what performance actually means, to the practical methodology of breaking individual pods to understand their limits, and navigating the complexities of Kubernetes-specific components that affect test results.

You will learn:

How to establish baseline performance metrics by systematically testing individual pods, disabling autoscaling features, and documenting each incremental change to understand real application limits

Why shared Kubernetes components skew results and how ingress controllers, service meshes, and monitoring stacks create testing challenges that require careful consideration of the entire request chain

Practical approaches to HPA configuration, including how to account for scaling latency, the time delays inherent in Kubernetes scaling operations, and planning for spare capacity based on your SLA requirements

The role of observability tools like OpenTelemetry in production environments where load testing isn't feasible, and how distributed tracing helps isolate performance bottlenecks across interdependent services

Sponsor

This episode is sponsored by Learnk8s — get started on your Kubernetes journey through comprehensive online, in-person or remote training.

More info

Find all the links and info for this episode here: https://ku.bz/yY-FnmGfH

Interested in sponsoring an episode? Learn more.

via KubeFM https://kube.fm

May 27, 2025 at 06:00AM

The Missing Link: How MCP Servers Supercharge Your AI Coding Assistant

Discover the power of Model Context Protocol (MCP) for AI-assisted software engineering! This video explores how MCP enhances Large Language Models and AI agents by providing crucial context. Learn about two essential MCP servers: Memory and Context7. See how they improve AI's ability to understand project specifics, retain information, and access up-to-date documentation. Witness practical demonstrations using Cursor, and learn how to integrate MCP servers into your workflow. Elevate your AI-assisted coding experience with MCP!

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Sponsor: Stacklok Toolhive 🔗 https://github.com/stacklok/toolhive ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

AIForDevelopers, #ModelContextProtocol, #LLMEnhancements

Consider joining the channel: https://www.youtube.com/c/devopstoolkit/join

▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/the-missing-link-how-mcp-servers-supercharge-your-ai-coding-assistant 🔗 Model Context Protocol: https://github.com/modelcontextprotocol 🎬 Outdated AI Responses? Context7 Solves LLMs' Biggest Flaw: https://youtu.be/F0MLnVgk4as

▬▬▬▬▬▬ 💰 Sponsorships 💰 ▬▬▬▬▬▬ If you are interested in sponsoring this channel, please visit https://devopstoolkit.live/sponsor for more information. Alternatively, feel free to contact me over Twitter or LinkedIn (see below).

▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social ➡ LinkedIn: https://www.linkedin.com/in/viktorfarcic/

▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬ 🎤 Podcast: https://www.devopsparadox.com/ 💬 Live streams: https://www.youtube.com/c/DevOpsParadox

▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬ 00:00 MCP Intro 00:50 What is Model Context Protocol (MCP)? 06:53 Memory and Context7 MCP Servers 07:49 Stacklok Toolhive (sponsor) 09:04 Memory and Context7 MCP Servers (cont.)

via YouTube https://www.youtube.com/watch?v=n0dCFY6wMeI