Why Kubernetes Discovery Sucks for AI (And How Vector DBs Fix It) DevOps & AI Toolkit 2025-09-08 | Discover why the Kubernetes API is brilliant for execution but a complete nightmare for discovery, and learn how semantic search with vector databases can finally solve this problem. This video demonstrates the real-world challenge of finding the right Kubernetes resources when you have hundreds of cryptically named resource types in your cluster, and shows how AI struggles with the same discovery issues that plague human users.We'll walk through a practical scenario where you need to create a PostgreSQL database with schema management in AWS, revealing how traditional keyword-based searching through 443+ Kubernetes resources becomes an exercise in frustration. Even when filtering by logical terms like "database," "postgresql," and "aws," the perfect solution remains hidden because it doesn't match your search keywords. The video then introduces a game-changing approach using vector databases and semantic search that enables both humans and AI to discover resources through natural language queries, regardless of exact keyword matches. By converting Kubernetes resource definitions into embeddings that capture semantic meaning, we transform an unsearchable cluster into an instantly discoverable one where you can simply describe what you want to accomplish rather than memorizing cryptic resource names.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: UpCloud🔗 upcloud.com▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#KubernetesAPI #SemanticSearch #VectorDatabaseConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/kubernetes/why-kubernetes-discovery-sucks-for-ai-and-how-vector-dbs-fix-it🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Kubernetes API Discovery with AI01:30 UpCloud (sponsor)02:37 Kubernetes API Discovery Nightmare11:33 Why AI Fails at Kubernetes Discovery16:47 Vector Database Semantic Search Solution23:15 Semantic Search Pros, Cons, and Key Takeaways
ElevenLabs API Review: A Developers Honest Take DevOps & AI Toolkit 2026-02-09 | A software engineer shares how AI addiction led to building a fully automated YouTube workflow, with a focus on integrating multimedia APIs into real applications. Rather than clicking through UIs like a "normie," he built a custom CLI that handles everything from brainstorming to publishing to dubbing videos into other languages—all through API calls.The video dives deep into ElevenLabs as a case study for audio and video APIs, demonstrating how dubbing that would normally require dozens of manual steps can be reduced to a single command. Beyond content creation, the discussion explores practical applications for DevOps and SRE engineers: localizing documentation, generating voice alerts for incidents, auto-narrating demos, and more. The honest review covers both the impressive API quality and voice output alongside real frustrations like slow support, broken YouTube URL dubbing, and concerning terms of service around voice data rights. For engineers looking to add multimedia capabilities to their workflows, this offers a practical, code-first perspective on what works and what to watch out for.#AI #SoftwareEngineering #ElevenLabsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/ai/elevenlabs-api-review-a-developers-brutally-honest-take➡ DevOps AI Toolkit en Español: youtube.com/@DevOpsAIToolkitSpain🔗 ElevenLabs: elevenlabs.io▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Dubbing with AI (ElevenLabs)01:45 ElevenLabs API for Developers08:04 Is ElevenLabs Worth It?
AI Agent Debugging Setup: OpenTelemetry + Jaeger in Kubernetes DevOps & AI Toolkit 2026-02-06 | This video tackles a critical challenge in AI agent development: understanding what's actually happening when agents behave unpredictably. Using a real-world example where two identical requests to the same AI agent produced dramatically different results—one taking 10 seconds with 10 operations, the other over a minute with 42 operations—the video demonstrates why observability is essential for agentic systems. Since LLMs decide their own execution paths, choosing which tools to call and how many times to loop, traditional debugging approaches fall short.The solution presented is OpenTelemetry tracing, the same technology used to monitor distributed systems, now applied to AI agents. The video explains how OTel's standardized `gen_ai.*` semantic conventions capture AI-specific telemetry across any model provider, while integrating seamlessly with existing infrastructure observability. Through a hands-on demonstration using Jaeger and a Kubernetes-based AI agent, viewers learn how to trace the complete journey of requests through agents, LLMs, tool executions, and external services. The key takeaway: while AI-specific tools like LangSmith have their place for prompt debugging and evaluations, OpenTelemetry provides a unified, vendor-neutral standard that connects AI traces to everything else in your stack—giving you the visibility needed to debug, optimize, and trust your agentic systems.#OpenTelemetry #AIAgents #ObservabilityConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/observability/ai-agent-debugging-setup-opentelemetry-jaeger-in-kubernetes🔗 OpenTelemetry: opentelemetry.io🎬 Distributed Tracing Explained: OpenTelemetry & Jaeger Tutorial: youtu.be/Oa-zqv-EBpw▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Tracing for AI Agents 01:04 Why AI Agents Need Tracing09:29 OTel vs AI-Specific Tools
DevOps Q&A: Helm Charts, Cilium Service Mesh, AI Tooling, and GitOps Promotion DevOps & AI Toolkit 2026-02-06 | In this AMA livestream, Viktor and Scott tackle a wide range of questions from the community covering platform engineering, Kubernetes tooling, and the evolving AI landscape. The conversation kicks off with thoughts on Claude's anti-ad campaign against OpenAI, then dives into practical topics like Helm chart management strategies, upgrading air-gapped environments using tools like the Carvel suite and Helm relocation utilities, and the Chainguard fork of Kaniko for container image builds. The duo also debates Cilium versus Istio for service mesh capabilities, noting Cilium's limitations around pod-to-pod mTLS on the same node, and shares their straightforward approaches to note-taking using Markdown files and GitHub issues.The session gets especially lively around environment promotion strategies, where both hosts advocate for simple YQ-based workflows over complex GitOps promotion tools that unnecessarily tie themselves to specific platforms like Argo CD. They discuss immutable container image promotion with cosign signing, the importance of building Kubernetes controllers using frameworks like Crossplane or Kro before writing custom ones, and trunk-based development versus coordinated multi-repo releases. The conversation wraps up with a deep dive into AI and platform engineering, where Viktor shares his work on data chunking for RAG embeddings, and both hosts emphasize that MCP servers and skills will become the primary interfaces for developer platforms—predicting that tools like Cline and Cursor will replace browser-based portals as the way developers interact with their platforms.▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro (skip to first question)08:48 Thoughts on Claude's stance against ads in AI12:31 Using Copier project for org base Helm charts14:22 Strategies for upgrading air-gapped environments18:19 Chainguard forking Ingress NGINX - thoughts?20:24 Good replacement for Kaniko after Google retirement?22:14 Migrating from Istio sidecar to Cilium for CNI/service mesh24:54 Opinion on Obsidian and note-taking for platform engineers28:04 Promotion strategies for immutable container images across registries31:39 Recommendation: try Ko for daemonless builds32:34 Developer pathway with right balance of abstraction34:38 Managing environment promotion with Argo CD and GitOps38:40 Where to draw the line exposing cluster config to users41:24 Using HTTPS for pod-to-pod communication inside cluster43:16 Controllers as boundaries - when to build your own49:19 Release strategy for multiple repos defining a stack51:26 One-person platform team supporting 250 devs - too early for Backstage?58:08 Is it hard being a contrarian about popular tools?1:03:44 HTTPS for pod-to-pod: manage own CA for local DNS?1:04:19 AI toolkit for platform engineering - where to start1:10:24 Running Terraform from GitHub Actions with least privilege on AWS
DevOps Q&A: MCP Servers, Kubernetes Observability, and Progressive Delivery DevOps & AI Toolkit 2026-01-30 | In this AMA session, Viktor and Scott dive into a wide range of DevOps and platform engineering topics submitted by their live audience. They explore the pros and cons of push versus pull-based observability models, discussing when Prometheus and OpenTelemetry each make sense, particularly in Kubernetes environments. The conversation covers the evolving role of Kubernetes engineers, with both hosts emphasizing that Kubernetes is becoming table stakes and the industry is shifting toward broader platform engineering responsibilities.The duo tackles questions about replacing OpenShift with alternatives like Spectro Cloud and Rancher, the value of certifications like Red Hat's RHCSA and the CKA, and feature flagging solutions built on the OpenFeature spec. They also discuss progressive delivery strategies using tools like Argo Rollouts, the relationship between service mesh and canary deployments, and practical tips for implementing Carpenter for node autoscaling. The session wraps up with insights on Backstage as an internal developer platform, the growing importance of MCP servers in agentic workflows, and recommendations for keeping internal components updated using Renovate or Dependabot.▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro (skip to first question)10:30 Push vs pull-based observability in Kubernetes18:00 Future of Kubernetes engineer jobs in 202622:47 Alternatives to OpenShift for bare metal clusters29:44 Is Red Hat RHCSA certification worth getting?32:00 Preferred feature flagging solution or platform35:44 Canary releases: Argo Rollouts vs Istio service mesh42:00 Using Agones for video game servers on Kubernetes42:40 Should I learn to develop a Kubernetes operator?43:51 Tips for adding Karpenter to a cluster48:59 Will building MCPs be required for DevOps/infra?54:40 OpenShift with HA Proxy for canary deployments56:18 Does progressive delivery help release safely?59:15 Key features that make Backstage stand out1:07:02 Migrating EC2 services to Kubernetes with KubeVirt1:10:11 Best technique to keep internal components updated
How I Built a Complete Developer Portal in Minutes With AI DevOps & AI Toolkit 2026-01-29 | This video demonstrates how to set up Port, a commercial internal developer portal, using AI and the Model Context Protocol (MCP) instead of spending weeks clicking through web UIs or writing endless JSON configurations. Starting from a completely empty Port account, the presenter shows how a coding agent like Claude Code can automatically discover the environment, create blueprints for Kubernetes resources and custom CRDs, deploy exporters through ArgoCD, configure GitHub integrations, and set up self-service actions—all through MCP tool calls.The video covers both the platform engineer perspective (setting up the entire portal infrastructure) and the platform user perspective (querying app status and creating resources interactively through the terminal). While highlighting Port's impressive MCP implementation with 27 tools covering blueprints, entities, scorecards, actions, and integrations, the presenter also discusses current limitations including large context size, lack of auto-discovery for Kubernetes resources, and missing tools like integration updates. The key takeaway is that Port's investment in their API and data lake architecture has positioned them well for the AI era, turning their portal into a powerful context source for coding agents.#DeveloperPortal #InternalDeveloperPlatform #AIAutomationConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/internal-developer-platforms/stop-setting-up-developer-portals-manually-feat-port-mcp🔗 Port: port.io🎬 How To Build A UI For An Internal Developer Platform (IDP) With Port?: youtu.be/ro-h7tsp0qI🎬 Mastering Developer Portals: Discover & Integrate API Schemas with Port: youtu.be/PV1sBiC85Yc▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Internal Developer Platforms with Port and AI 01:36 Port for Platform Engineers05:59 Port for Platform Users07:31 The Prompt Behind It08:08 Port MCP: The Good and Bad
Containers Are NOT Virtual Machines DevOps & AI Toolkit 2026-01-23 | Containers Are NOT Virtual MachinesWatch the full video: youtu.be/ueTe-VQaD7c#Shorts
DevOps AMA DevOps & AI Toolkit 2026-01-23 | Ask us anything about DevOps, Cloud, Kubernetes, AI, GitOps, SRE, or anything else.▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
My Production Dockerfile Rules: How I Build Docker Images DevOps & AI Toolkit 2026-01-22 | This video tackles a critical but often overlooked aspect of containerization: writing secure, optimized Dockerfiles for production environments. It exposes common mistakes found in most production Dockerfiles—running as root, using latest tags, copying entire directories with secrets, and bloated images filled with unnecessary debugging tools—then systematically walks through every best practice needed to fix them.The tutorial covers four key areas: choosing minimal base images and implementing multi-stage builds, optimizing layer caching and build efficiency, hardening security through non-root users and pinned versions, and maintaining clean, well-documented configurations. Each principle is demonstrated with practical examples from a real Docusaurus project. The video concludes with a powerful demonstration of how AI-powered tools can automatically analyze a project and generate production-ready Dockerfiles that incorporate all these best practices, transforming a tedious manual process into something that takes seconds while producing smaller, faster, and more secure container images.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: MetalBear🔗 metalbear.com▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#Docker #ContainerSecurity #DevOpsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/containers/my-production-dockerfile-rules-how-i-build-docker-images🔗 DevOps AI Toolkit: devopstoolkit.ai▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Docker Best Practices 00:52 mirrod by MetalBear (sponsor)02:13 Dockerfile Base Images05:29 Dockerfile Layer Caching08:30 Dockerfile Security Hardening12:52 Dockerfile Maintainability14:55 AI Dockerfile Generator
DevOps AMA: AI Agents, External Secrets, and Career Advice DevOps & AI Toolkit 2026-01-16 | In this solo AMA session, Viktor fields a wide range of questions from the DevOps and platform engineering community. The discussion covers practical tool comparisons like External Secrets versus Vault agents, with Viktor expressing his preference for solution-agnostic approaches. He breaks down the distinction between SRE and DevOps roles, explaining how AI fits differently into each—SREs using AI to help with operational analysis and remediation, while DevOps engineers focus on building AI-powered interfaces and agents for their users.The conversation dives deep into AI tooling recommendations, with Viktor enthusiastically endorsing Claude Code for terminal-based development and Code Rabbit for automated PR reviews. He shares insights on agent frameworks, suggesting Vercel's AI SDK for serious development work and K agent for quick prototyping, while cautioning about the limitations of each. Viktor also discusses his experimental work on dynamic AI-driven UIs that can render unpredictable outputs based on natural language queries, moving away from traditional static dashboards.Additional topics include GitOps considerations for ephemeral environments, getting started with Kyverno for Kubernetes policy management, Crossplane for infrastructure control planes, and essential advice for first-time KubeCon attendees. Viktor addresses concerns about AI replacing engineers, arguing that while short-term disruption is real, the long-term trend will see the same number of people doing more work rather than fewer people doing the same work—emphasizing that those who embrace AI tools will be best positioned to thrive.▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro (skip to first question)06:22 External Secrets vs Vault Agent for secrets management07:52 What should DevOps/SRE learn about AI?11:51 Simplest and cheapest way to deploy website with backend and database13:12 Migrating Java EAP to OpenShift advice15:32 Does Upbound platform support operators other than Crossplane?16:55 Getting started with Kyverno on EKS18:51 Best framework for creating deployment agents21:11 GitOps for ephemeral preview environments - Argo CD PR generator26:08 Getting started with Kyverno - is it safe to just start?27:07 Tool for automated PR review approval for Terraform/Atlantis28:55 Thoughts on ToolHive for MCPs32:04 Most interesting tool you've used in 202536:33 What type of agents can DevOps provide to dev teams?39:40 Can static portals handle fast-changing APIs or need AI UI?44:46 Should DevOps engineer with 4 years experience switch to AI?46:34 Code Rabbit review tool discussion47:52 Tips for first KubeCon EU attendance51:57 Differences between US and EU KubeCon53:46 Core tools to know for breaking into DevOps/SRE role55:46 Prediction for AI replacing platform engineers59:01 Is DevOps AI-proof?1:01:57 Future of job security in today's AI world1:03:10 What skills should DevOps engineers focus on to stay relevant?1:04:05 Thoughts on using Crossplane as central control plane1:04:38 MacBook Pro recommendation1:05:05 Where are you from?1:05:46 CEO misconceptions about AI solving everything1:11:26 When should operations engineers start writing clean code?1:12:50 Thoughts on Microsoft Foundry (Azure Foundry)
API Gateways Explained: Why Your Services Are a Mess (Zuplo Review) DevOps & AI Toolkit 2026-01-15 | API gateways solve a universal problem: every API you expose needs authentication, rate limiting, documentation, and analytics, but implementing these in every service creates scattered logic and inconsistent policies. This video breaks down what API gateways actually are, when they're essential, and when they're overkill. You'll learn how gateways fit into different architectures—from traditional VMs to Kubernetes to serverless—and why they're critical for north-south traffic but completely unnecessary for service-to-service communication.The video explores Zuplo as a practical example, walking through both their API Management gateway for REST APIs and their AI Gateway for LLM traffic. You'll see how to import OpenAPI schemas, attach policies like rate limiting, and integrate with GitOps workflows. For AI applications, the gateway provides centralized cost tracking, provider failover, and unified API keys across your organization. The honest review covers both strengths (excellent developer experience, edge deployment, built-in monetization) and limitations (schema duplication, TypeScript-only custom logic, no self-hosting option) to help you decide if it's the right fit for your needs.#APIGateway #Zuplo #DevOpsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/development/api-gateways-explained-why-your-services-are-a-mess-zuplo-review🔗 Zuplo: zuplo.com▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 API Gateways & Zuplo00:55 API Gateway Explained06:23 Zuplo API Management16:04 Zuplo AI Gateway20:41 Zuplo Review
Why Your Platform Needs AI Agents DevOps & AI Toolkit 2026-01-10 | Why Your Platform Needs AI AgentsWatch the full video: youtu.be/65o_j4E7_lk#Shorts
DevOps Q&A: AI Workflows, Kubernetes Cost Optimization, and MCP Servers DevOps & AI Toolkit 2026-01-09 | In this AMA session, Viktor and Scott dive into a wide range of topics spanning AI workflows, Kubernetes operations, and platform engineering. They share their personal approaches to prompt engineering and context management when working with AI coding assistants, emphasizing the importance of keeping tasks small and managing context windows effectively to avoid hallucinations. The discussion explores how AI is transforming operations work, with particular focus on bridging knowledge gaps for traditional infrastructure teams and the challenges of feeding the right data to AI systems.The conversation covers practical DevOps concerns, including real-time alerting strategies in Kubernetes, comparing push-based OpenTelemetry approaches with pull-based Prometheus models. Viktor and Scott also discuss Kubernetes cost optimization, recommending starting with node autoscaling tools like Karpenter before tackling workload right-sizing. They weigh in on building Kubernetes operators, strongly advocating for tools like Crossplane over custom operators when possible, and share thoughts on the current AI landscape, acknowledging a bubble while emphasizing that AI remains genuinely useful when implemented correctly with proper context engineering.▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro (skip to first question)06:29 What's your AI workflow with specs and tools?18:25 Who should own DevEx in an organization?21:21 Real-time alerting approach for Kubernetes workloads29:07 Which part of ops gains most from AI?40:40 Where to start writing a Kubernetes operator?46:26 Low-hanging fruit for Kubernetes cost optimization51:24 Changes in Upbound plans explained51:57 Is there an AI bubble?59:04 Crossplane composites vs Flux resource sets1:01:35 Using on-prem open models vs cloud LLMs for PRs
Dont Lock Your Agents to One Model DevOps & AI Toolkit 2026-01-09 | Don't Lock Your Agents to One ModelWatch the full video: youtu.be/65o_j4E7_lk#Shorts
Why Ops Teams Are Behind on AI DevOps & AI Toolkit 2026-01-08 | Why Ops Teams Are Behind on AIWatch the full video: youtu.be/65o_j4E7_lk#Shorts
AI Code Review Is a No-Brainer DevOps & AI Toolkit 2026-01-07 | AI Code Review Is a No-BrainerWatch the full video: youtu.be/65o_j4E7_lk#Shorts
Your Portal Is NOT a Platform DevOps & AI Toolkit 2026-01-06 | Your Portal Is NOT a PlatformWatch the full video: youtu.be/65o_j4E7_lk#Shorts
Top 10 DevOps & AI Tools You MUST Use in 2026 DevOps & AI Toolkit 2026-01-05 | This video presents a practitioner's guide to the most essential developer tools for 2026, covering both the AI tools and the foundational technologies that remain critical. Rather than offering a neutral comparison, it shares battle-tested recommendations based on months of real-world use across AI models, coding agents, custom agent development, code review automation, vector databases, internal developer platforms, Kubernetes development environments, platform testing, and modern shell scripting.Key recommendations include Anthropic's Claude for AI-powered software engineering, Cursor or Claude Code for coding agents depending on your workflow preference, Vercel AI SDK for building custom agents with model flexibility, CodeRabbit for automated code reviews with MCP integration, Qdrant for vector database needs, the BACK Stack for building internal developer platforms on Kubernetes, mirrord for bridging local and remote development environments, Kyverno Chainsaw for declarative platform testing, and Nushell for modern scripting with structured data handling. The video emphasizes that while agentic AI has transformed how developers work, solid foundations like testing frameworks, development environments, and platform architecture still matter—AI now intersects with all of them rather than replacing them.#DevOps #AITools #KubernetesConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/devops/top-10-devops-tools-you-must-use-in-2026▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 DevOps and AI Tools 202602:10 Best AI Models for Software Engineering06:19 Best AI Coding Agents11:45 Building Custom AI Agents17:15 AI Code Review Tools21:09 Vector Databases for AI25:06 Internal Developer Platforms30:26 Kubernetes Dev Environments34:56 Kubernetes Platform Testing39:31 Modern Shell Scripting42:46 What to Use in 2026
Kubernetes ownerReferences Explained: Why Your Apps Are a Mess DevOps & AI Toolkit 2025-12-29 | Ever wondered why `kubectl get all` doesn't actually get all your resources? It conveniently ignores Ingresses, PersistentVolumeClaims, and many other resource types. Even worse, when you do list everything in a namespace, you're left staring at a pile of disconnected objects with no way to understand how they relate to each other or form complete systems. This video dives into this fundamental Kubernetes problem and explores how ownerReferences and resource hierarchies actually work under the hood.To solve this challenge, I built a custom Solution CRD that wraps related resources into logical groups with clear context, intent, and aggregated status. Instead of manually piecing together which Deployments, Services, Ingresses, and databases belong to the same application, you can now define and query complete solutions as first-class citizens in your cluster. I'll walk you through the problem, demonstrate tools like kubectl-tree for exploring ownership hierarchies, and show you how this simple CRD approach finally answers questions like "What is this app?" and "Is my entire system healthy?" Check out the project at github.com/vfarcic/dot-ai-controller if you want to try it yourself.#Kubernetes #CRD #DevOpsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/kubernetes/stop-trusting-kubectl-get-all-heres-what-it-hides-from-you🔗 DevOps AI Toolkit Controller: github.com/vfarcic/dot-ai-controller▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Kubernetes Resource Relations01:13 What Is This Thing in Kubernetes?04:50 Kubernetes ownerReferences and Garbage Collection08:30 Solving Resource Grouping with CRDs
Stop Resisting AI or Get Left Behind! (A Wake-Up Call) DevOps & AI Toolkit 2025-12-22 | What happens when the team that always adapts first suddenly refuses to play by new rules? This video tells the story of a seemingly safe bet that went terribly wrong—a bet on a team with every advantage imaginable: the most skilled players, unlimited budgets, deep experience, and even the power to make the rules. But when AI was allowed on the field, everything changed. While other teams embraced the new reality and showed up with full rosters ready to play, that team's players mostly refused to participate, calling it hype and yelling at the few teammates who dared to try.This story is about what's happening right now in tech companies with AI. The teams that led every previous transformation—VMs, cloud, containers, Kubernetes—are sitting on the sidelines while historically resistant teams are running full speed ahead. The irony is brutal, and the lesson is clear: being adaptable in the past doesn't guarantee you'll adapt in the future. The question isn't whether AI will change how we work, but whether you'll be in the field playing or on the bench yelling at those who are.#AIAdoption #TechLeadership #ChangeManagementConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/ai/stop-resisting-ai-or-get-left-behind-a-wake-up-call▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Distributed Tracing Explained: OpenTelemetry & Jaeger Tutorial DevOps & AI Toolkit 2025-12-17 | Your users are complaining about slow response times—sometimes 8 seconds, other times 2 seconds—but your metrics show everything is fine. Average response times look acceptable, all services report healthy, and your dashboards are green. So what's really happening? The problem is that what looks like a single user request is actually dozens of separate, independent requests cascading through your microservices. Each service only sees its own operations, with no way to know they're part of the same logical transaction. Your logs show individual services completed successfully, but you can't correlate these entries across services or identify which specific operation is causing the delay.This video shows you exactly how to solve this blindness using distributed tracing with OpenTelemetry. You'll learn the difference between automatic and manual instrumentation, see real examples of tracing implementation in TypeScript, and analyze actual traces using Jaeger to understand request flows through complex systems. We'll cover traces, spans, context propagation, semantic conventions, sampling strategies, and how to export trace data to any backend without vendor lock-in. By the end, you'll understand why traditional observability tools can't see what's happening in distributed systems and how to implement tracing that reveals the complete journey of every request through your architecture.#DistributedTracing #OpenTelemetry #Microservices▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: DevStats🔗 https://devstats.plug.dev/5W1oh9J▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Consider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/observability/distributed-tracing-explained-opentelemetry--jaeger-tutorial🔗 OpenTelemetry: opentelemetry.io▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Distributed Tracing with OpenTelemetry (OTEL)01:18 DevStats (sponsor)02:34 Microservices Performance Mystery06:24 OpenTelemetry Distributed Tracing10:57 Analyzing Traces with Jaeger14:53 Understanding OpenTelemetry Traces20:14 Tracing Solves Observability Blindness
DevOps Q&A: Kubernetes Essentials, Learning Strategies, and Gateway API Migration DevOps & AI Toolkit 2025-12-12 | In this AMA livestream, Viktor and Scott dive into essential Kubernetes topics, starting with their top five must-have components for a state-of-the-art cluster. Their recommendations include Kyverno for policies, GitOps tools like Argo CD or Flux, Crossplane for building platform abstractions, External Secrets Operator, External DNS, and CNPG for self-hosted PostgreSQL databases. The hosts also share their approaches to continuous learning, emphasizing hands-on experimentation, listening to conference talks, and quickly evaluating new tools based on their quick-start experience.The conversation covers practical advice for junior DevOps engineers, including the importance of learning beyond company-specific practices and exploring alternative tools at home. Viktor and Scott discuss the value of communication skills in engineering careers, the challenges of migrating from NGINX Ingress to Gateway API, and the ongoing debate between Terraform and Crossplane for infrastructure management. They also touch on Backstage as an internal developer portal, the significance of API-driven architectures, and the concept behind the CNCF "Bactstack" combination of tools. Throughout the session, both hosts stress the importance of understanding fundamentals, being open to changing your mind when presented with better arguments, and continuously challenging your own knowledge.▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro05:25 Must-have Kubernetes controllers and components?08:31 How do you continue learning new tech?16:02 Advice for junior DevOps engineer feeling lost20:50 Project that would make you job-ready?22:09 Can a law student leverage degree in tech?25:32 Is communication more important than technical skills?31:34 When did you feel you really know stuff?38:50 Is Backstage good for Terraform deployments?42:16 Should we switch from Terraform to Crossplane?52:10 Where to migrate from Ingress NGINX?57:19 Best Ingress NGINX alternative?57:42 Do you still use VKS for on-prem Kubernetes?60:05 Any thoughts on Costack?60:18 Do you use specific AI prompting techniques?63:46 Is Kubernetes the only GitOps backend?64:46 Is BMED good for AI agent task planning?
Top 10 GitHub Features You MUST Use in 2026! DevOps & AI Toolkit 2025-12-08 | Tired of chaotic repositories with vague bug reports, unclear pull requests, and outdated dependencies? This video shows you how to transform any GitHub repository into a professionally organized project using built-in tools most developers don't even know exist. Learn how to set up issue templates that force useful information, pull request templates that make code review actually possible, automated dependency updates with Renovate, security scanning with OpenSSF Scorecard, automatic PR labeling, stale issue management, and all the essential governance files that make collaboration smooth. Whether you're maintaining open source projects, building internal tools, or managing commercial software, these GitHub features will save you hours of frustration.The best part? You don't have to spend hours setting this up manually. The video demonstrates how to automate the entire process in minutes using the DevOps AI Toolkit's Project Setup MCP tool with Claude Code. See real examples from the DevOps AI Toolkit repository itself, including how coding agents can automatically fill out PR templates, how automated workflows handle dependencies and security scanning, and how proper documentation and governance files prevent confusion and establish clear processes for contributors. By the end, you'll know exactly how to create a professional repository that works for both human contributors and AI coding agents.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: JFrog Fly🔗 jfrog.com/fly_viktor▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#GitHubTemplates #DevOpsAutomation #ProjectManagementConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/development/top-10-github-project-setup-tricks-you-must-use-in-2025🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai🎬 How I Tamed Chaotic AI Coding with Simple Workflow Commands: youtu.be/LUFJuj1yIik▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 GitHub Setup02:03 JFrog Fly (sponsor)03:13 GitHub Issue Templates08:13 Pull Request Templates12:26 Code Owners and Auto-Assignment15:25 Security Scanning and Badges16:50 Automated Dependency Updates21:12 Essential Documentation Files24:17 DevOps AI Toolkit
DevOps Q&A: GitOps Tools, Backstage Setup, and Platform Engineering DevOps & AI Toolkit 2025-12-05 | In this AMA livestream, Viktor and Scott dive into a wide range of DevOps and platform engineering topics submitted by viewers. The discussion covers practical guidance on Kubernetes networking, including recommendations to migrate from the soon-to-be-discontinued Ingress NGINX to Gateway API for improved flexibility and throughput management. They explore fleet management strategies for organizations running 40+ clusters, suggesting combinations of Flux, Crossplane, and dedicated fleet management tools.The hosts share insights on implementing Backstage as an internal developer portal, emphasizing the importance of auto-ingesting resources, using tech insight scorecards to gamify adoption, and adapting organizations to tools rather than the reverse. They also discuss MCP (Model Context Protocol) gateways, highlighting the need for better authentication and tool selection capabilities. Other topics include ephemeral development environments, the differences between Tekton and Argo CD, Cilium deployment best practices, and using Crossplane for on-premises infrastructure. Throughout the session, Viktor and Scott stress the importance of using AI for non-repeatable tasks while relying on automation for predictable workflows.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Intro (skip to first question)06:52 Best way to compare policies using AI?09:44 Different ingress controllers for streaming vs API calls?12:31 Fleet management approach for 40+ clusters with Flux/Crossplane?15:00 Tips for implementing Backstage for software inventory and governance?18:55 Buffering concerns with multipart uploads on NGINX ingress23:07 Good MCP gateway recommendations for multiple MCP servers?26:36 Thoughts on Flux Uncontained?27:47 Deploy Cilium via Argo CD or before cluster creation?28:46 Is standardizing components a priority for building an IDP?32:47 Is Crossplane worth using for on-prem infrastructure?36:20 Any trends or ideas from KubeCon?43:07 Public repo for Backstage Crossplane workshop?44:14 Advice on mapping org structure to Backstage data model?50:08 Best practices for adapting to tools vs adapting tools to you55:37 Is Flox a viable container alternative?58:05 Thoughts on ephemeral environments and DevPod for IDP?1:02:24 Difference between Argo CD and Tekton - when to use each?1:05:25 Integrating Backstage with GitOps across multiple repos?
Ep41 - Ask Me Anything About Anything with Scott Rosenberg 📱 DevOps & AI Toolkit 2025-12-05 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Deploy AI Agents and MCPs to Kubernetes: Is kagent and kmcp Worth It? DevOps & AI Toolkit 2025-12-01 | This video explores kagent and kmcp, two tools that promise to bring AI agents and MCP servers into Kubernetes using cloud-native principles. kagent lets you define AI agents as custom resources with YAML manifests, connect them to MCP servers for tools, and manage them like any other Kubernetes workload. kmcp deploys MCP servers to Kubernetes clusters using simple custom resources. The concept sounds appealing for platform engineers: create agents declaratively, give them specific tools, let them communicate through open protocols like A2A, all running in your existing infrastructure.However, the reality reveals significant gaps. While kagent successfully deploys agents to Kubernetes and connects them to MCP tools, its web interface is severely lacking compared to modern coding agents like Claude Code or Cursor. Tool execution is unreliable, there's no built-in user confirmation before calling tools, and the choice to expose agents via the A2A protocol instead of MCP limits integration with existing coding tools. kmcp works for deploying MCP servers but offers limited value beyond what standard Kubernetes manifests or Helm charts already provide. The video demonstrates both tools in action—creating agents, connecting to MCP servers, and troubleshooting Kubernetes issues—while honestly examining whether these projects solve real problems or just add unnecessary complexity to workflows that modern coding agents already handle better.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: RavenDB🔗 Meet the new AI Agent in RavenDB: ravendb.net/ai-agent-creator?utm_source=youtube&utm_medium=influencers&utm_campaign=devops_toolkit🔗 Visit RavenDB's homepage: ravendb.net/?utm_source=youtube&utm_medium=influencers&utm_campaign=devops_toolkit▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#KubernetesAI #MCPServers #AIAgentsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/kubernetes/deploy-ai-agents-and-mcps-to-k8s-is-kagent-and-kmcp-worth-it🔗 kagent: https://kagent.dev▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 AI Agent and MCPs in Kubernetes01:01 RavenDB (sponsor)02:25 Kubernetes AI Agents with kagent11:42 Integrating External MCP Servers16:23 Deploying MCP Servers with kmcp22:31 Should You Use kagent and kmcp?
Ep40 - Ask Me Anything About Anything with Scott Rosenberg 📱 DevOps & AI Toolkit 2025-11-28 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Ep40 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-11-28 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Why Gemini 3 Feels Like Pair Programming With a Grumpy Coder DevOps & AI Toolkit 2025-11-24 | Gemini 3 is undeniably fast and impressive on benchmarks, but after a full week of real-world software engineering work, the reality is more complicated. While everyone's been hyping its capabilities based on day-one reviews and marketing materials, this video digs into what actually matters: how Gemini 3 performs with coding agents on real projects, not just one-shot Tetris games or simple websites. The speed is remarkable at 128 tokens per second, but it comes with serious trade-offs that affect daily pair programming work.The core issues are frustrating: Gemini 3 is nearly impossible to redirect once it commits to a plan, suffers from an 88% hallucination rate (nearly double Sonnet 4.5's 48%), and confidently claims tasks are complete when they're not. It ignores context from earlier in conversations, struggles with complex multi-step instructions, and dismisses suggestions like a grumpy coder who thinks they know best. While it excels at one-shot code generation, it falls short as a collaborative partner for serious software development. Gemini 3 is genuinely one of the best models available (probably second place behind Sonnet 4.5) but it's not the massive leap forward that the hype suggests, and the gap between Claude Code and Gemini CLI remains significant.#Gemini3 #AIcoding #SoftwareEngineeringConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/ai/gemini-3-is-fast-but-gaslights-you-at-128-tokens-second🔗 Gemini 3: https://deepmind.google/models/gemini▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Gamini 3 with Gamini CLI00:25 Gemini 3 Real-World Testing02:54 Gemini 3's Biggest Problems10:10 Is Gemini 3 Worth It?
Ep39 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-11-21 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Ep39 - Ask Me Anything About Anything with Scott Rosenberg 📱 DevOps & AI Toolkit 2025-11-21 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
My Kubernetes Workflow: How AI Detects & Fixes Issues Faster DevOps & AI Toolkit 2025-11-17 | Tired of waking up at 3 AM to troubleshoot Kubernetes issues? This video shows you how to automate the entire incident response process using AI-powered remediation. We walk through the traditional manual troubleshooting workflow—detecting issues through kubectl events, analyzing pods and their controllers, identifying root causes, and validating fixes—then demonstrate how AI agents can handle all four phases automatically. Using the open-source DevOps AI Toolkit with the Model Context Protocol (MCP) and a custom Kubernetes controller, you'll see how AI can detect failing pods, analyze the root cause (like a missing PersistentVolumeClaim), suggest remediation, and validate that the fix worked, all while you stay in bed.The video breaks down the complete architecture, showing how a Kubernetes controller monitors events defined in RemediationPolicy resources, triggers the MCP server for analysis, and either automatically applies fixes or sends Slack notifications for manual approval based on confidence thresholds and risk levels. You'll learn how the MCP agent loops with an LLM using read-only tools to gather data and analyze issues, while keeping write operations isolated and requiring explicit approval. Whether you want fully automated remediation for low-risk issues or human-in-the-loop approval for everything, this approach gives you intelligent troubleshooting that scales beyond what you can predict and prepare for manually.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: JFrog Fly🔗 jfrog.com/fly_viktor▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#Kubernetes #AIAutomation #DevOpsConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/ai/ai-vs-manual-kubernetes-troubleshooting-showdown-2025🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Kubernetes Analysis and Remediation with AI01:15 JFrog Fly (sponsor)02:46 Kubernetes Troubleshooting Manual Process11:37 AI-Powered Kubernetes Remediation14:38 MCP Architecture and Controller Design20:49 Key Takeaways and Next Steps
AI Agent Architecture Explained: LLMs, Context & Tool Execution DevOps & AI Toolkit 2025-11-10 | You type "Create a PostgreSQL database in AWS" into Claude Code or Cursor, and it just works. But how? Most people think the AI does everything, but that's wrong. The AI can't touch your files or run commands on its own. This video breaks down the real architecture behind AI coding agents, explaining the three key players that make it all work: you (providing intent), the agent (the orchestrator), and the LLM (the reasoning brain). Understanding this matters if you're using these tools every day.We'll walk through increasingly sophisticated architectures, from basic system prompts to the complete agent loop that enables real work. You'll learn how tools get executed, what context really means, how the agent manages the loop between you and the LLM, and why the LLM is stateless. We'll also cover practical considerations like MCP (Model Context Protocol) for integrating external tools and context limits that affect performance. By the end, you'll understand that the agent is actually the only "dumb" actor in the system—it's pure execution with no intelligence. The LLM provides the brains, you provide the intent, and the agent coordinates everything to make it happen.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: RavenDB🔗 ravendb.net▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#AIAgents #LLM #HowAIWorksConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬➡ Transcript and commands: https://devopstoolkit.live/ai/ai-agent-architecture-explained-llms,-context--tool-execution▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 AI Agents Explained01:02 RavenDB (sponsor)02:16 How Do Agents Work?05:42 How AI Agent Loops Work?09:36 MCP (Model Context Protocol) & Context Limits11:38 AI Agents Explained: Key Takeaways
Which AI Model is Best for DevOps? I Tested 10 (Shocking Results) DevOps & AI Toolkit 2025-11-03 | A comprehensive, data-driven comparison of 10 leading large language models (LLMs) from Google, Anthropic, OpenAI, xAI, DeepSeek, and Mistral, specifically tested for DevOps, SRE, and platform engineering workflows. Instead of relying on traditional benchmarks or marketing claims, this evaluation runs real agent workflows through production scenarios: Kubernetes operations, cluster analysis, policy generation, manifest creation, and systematic troubleshooting—all with actual timeout constraints. The results reveal shocking gaps between benchmark promises and production reality: 70% of models couldn't complete tasks in reasonable timeframes, premium "reasoning" models failed on tasks cheaper alternatives handled easily, and the most expensive model ($120 per million output tokens) failed more tests than it passed.The evaluation measures five key dimensions: overall performance quality, reliability and completion rates, consistency across different tasks, cost-performance value, and context window efficiency. Five distinct test scenarios push models through endurance tests (100+ consecutive interactions), rapid pattern recognition (5-minute workflows), comprehensive policy compliance analysis, extreme context pressure (100,000+ token loads), and systematic investigation loops requiring intelligent troubleshooting. The rankings reveal clear performance tiers, with Claude Haiku emerging as the overall winner for its exceptional efficiency and price-performance ratio, while Claude Sonnet takes the reliability crown with 98% completion rates. The video provides specific recommendations on which models to use, which to avoid, and why cost doesn't always correlate with capability in production environments.#LLMComparison #DevOps #AIforEngineersConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/best-ai-models-for-devops--sre-real-world-agent-testing🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai🎬 Analysis report: github.com/vfarcic/dot-ai/blob/main/eval/analysis/platform/synthesis-report.md▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Large Language Models (LLMs) Compared01:54 How I Compare Large Language Models05:01 LLM Evaluation Criteria and Test Scenarios13:23 AI Model Benchmark Results27:34 AI Model Rankings and Recommendations
Ep38 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-10-31 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Build Self-Healing Kubernetes Systems With AI & Event Automation DevOps & AI Toolkit 2025-10-27 | Tired of being woken up at 2 AM to manually troubleshoot Kubernetes incidents that could be fixed automatically? This video explores how to build intelligent self-healing systems that watch Kubernetes events, analyze problems, and remediate issues before they ruin your weekend. We'll break down the complete automation pipeline—from understanding how Kubernetes events work and what makes them ideal triggers, to implementing a maturity progression from manual firefighting through rule-based automation to AI-assisted remediation.Learn when traditional automation works best (alerting and known patterns), where AI genuinely excels (analysis and unknown scenarios), and how to strategically combine both approaches. We'll cover the three phases of incident response—alerting, analysis, and remediation—and show you how to build systems that handle knowns with efficient controllers while leveraging AI for novel problems. The key is creating feedback loops that continuously graduate unknowns into automated knowns, progressively shrinking the surface area where human intervention is needed. Includes links to open-source projects demonstrating these principles in production.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: JFrog Fly🔗 jfrog.com/fly_viktor▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#Kubernetes #SelfHealingSystems #AIAutomationConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/kubernetes/self-healing-kubernetes-when-to-use-ai-vs-traditional-automation🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Kubernetes Remediation01:15 JFrog fly (sponsor) 02:43 Kubernetes Events Explained06:21 Kubernetes Automation Pipeline12:46 AI-Powered Kubernetes Remediation19:26 Building Self-Healing Systems
Ep37 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-10-24 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
How to Run MCP Servers: Docker vs Kubernetes vs Cloud Platforms DevOps & AI Toolkit 2025-10-20 | Discover the four main ways to deploy MCP servers, from simple local execution to enterprise-ready Kubernetes clusters. This comprehensive guide explores the trade-offs between NPX local deployment, Docker containerization, Kubernetes production setups, and cloud platform alternatives like Fly.io and Cloudflare Workers.You'll see practical demonstrations of each approach using a real MCP server, learning about security implications, scalability challenges, and team collaboration benefits. The video covers why local NPX execution creates security risks and dependency nightmares, how Docker provides better isolation but remains single-user, and why Kubernetes offers the best solution for shared organizational infrastructure. We also examine the ToolHive operator's limitations and explore various cloud deployment options with their respective vendor lock-in considerations. Whether you're developing MCP servers or deploying them for your team, this guide will help you choose the right deployment strategy for your specific needs.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Browserbase🔗 browserbase.com▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#MCP #ModelContextProtocolConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/mcp-server-deployment-guide-from-local-to-production🔗 Model Context Protocol: modelcontextprotocol.io▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Model Context Protocol (MCP) Deployment01:40 Browserbase (sponsor)02:50 MCP Local NPX Deployment06:32 MCP Docker Container Deployment09:23 MCP Kubernetes Production Deployment14:09 MCP ToolHive Kubernetes Operator19:15 Alternative MCP Deployment Options22:46 Choosing the Right MCP Deployment
Why Your Infrastructure AI Sucks (And How to Fix It) DevOps & AI Toolkit 2025-10-13 | Discover why your AI agent is completely failing at infrastructure management and learn to build an AI-powered Internal Developer Platform that actually works. Most organizations are treating AI like a search engine, asking vague questions and getting generic answers that break in production. This video reveals the five critical components that transform useless AI into intelligent infrastructure automation.You'll learn to build capabilities discovery using Vector databases for semantic search across Kubernetes resources, capture organizational patterns from tribal knowledge and documentation, create enforceable policies that guide AI toward compliance, implement proper context management to avoid the bloated mess most systems become, and design intelligent workflows that guide users to the right solutions instead of relying on guesswork. Watch as we demonstrate the complete transformation from a generic AI response to a fully functional PostgreSQL deployment that follows organizational patterns, enforces compliance policies, and deploys correctly the first time.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Tuple🔗 https://tuple.app/DOT👉 Promo code: DOT2025▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#AIInfrastructure #InternalDeveloperPlatform #KubernetesAIConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/internal-developer-platforms/why-your-infrastructure-ai-sucks-and-how-to-fix-it🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai🎬 Stop Blaming AI: Vector DBs + RAG = Game Changer: youtu.be/zqpJr1qZhTg🎬 Why Kubernetes Discovery Sucks for AI (And How Vector DBs Fix It): youtu.be/MSNstHj4rmk▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 AI for Infrastructure Challenges01:42 Tuple (sponsor)03:16 Why Your AI Agent Is Useless09:52 Kubernetes API Discovery That Actually Works13:41 Organizational Knowledge AI Can Actually Use17:49 Stop Breaking Production With AI22:17 The Context Window Disaster Nobody Talks About25:16 Smart Conversations That Get Results29:34 Your Complete AI-Powered IDP Blueprint
Ep36 - Ask Me Anything About Anything DevOps & AI Toolkit 2025-10-10 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Kubernetes Controllers Deep Dive: How They Really Work DevOps & AI Toolkit 2025-10-06 | Most people using Kubernetes know how to write YAML and run kubectl apply, but when things break, they're completely lost. The secret they're missing? Understanding controllers - the beating heart that makes Kubernetes actually work. Controllers are what automatically restart your crashed pods, scale your applications, and make custom resources feel native to the platform.This video dives deep into the real mechanics of how Kubernetes controllers operate. You'll discover how controllers consume and emit events to coordinate with each other, how the reconciliation loop continuously maintains your desired state, and how the Watch API efficiently streams changes without overwhelming the system. We'll explore custom resource definitions that extend Kubernetes, controller communication patterns, and the event-driven architecture that makes everything self-healing. Whether you're debugging cluster issues or building your own controllers, this knowledge will transform how you think about Kubernetes from just throwing YAML at the wall to truly understanding the orchestration engine underneath.#KubernetesControllers #Kubernetes #DevOpsEngineeringConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/kubernetes/kubernetes-controllers-deep-dive-how-they-really-work🔗 Kubernetes: kubernetes.io▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Kubernetes Controllers Deep Dive01:18 Kubernetes Control Loops Explained04:12 How Kubernetes Controllers Watch Events07:35 Kubernetes Event Emission11:56 Kubernetes Reconciliation Loop17:12 Kubernetes Watch API21:01 Kubernetes Custom Resource Definitions (CRDs)21:13 Kubernetes Controller Communication25:22 Kubernetes Controllers Mastery
Ep35 - Ask Me Anything About Anything DevOps & AI Toolkit 2025-10-03 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Octopus🔗 Enterprise Support for Argo: octopus.com/support/enterprise-argo-support▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
How I Tamed Chaotic AI Coding with Simple Workflow Commands DevOps & AI Toolkit 2025-09-29 | Tired of AI coding agents that jump between tasks chaotically and lose track of context? This video demonstrates a complete systematic workflow for AI-assisted development that keeps both you and your AI agent focused and organized from initial idea through production deployment.I'll walk you through my entire PRD-based development system, showing real implementation of a complex feature from start to finish. You'll see how to create comprehensive technical requirements with AI analysis, track progress systematically, handle inevitable plan changes, prioritize tasks intelligently, and complete features with full traceability. The workflow uses simple MCP commands like `/prd-create`, `/prd-next`, `/prd-update-progress`, and `/prd-done` to guide systematic development without requiring complex external tools. By the end, you'll understand how to transform chaotic AI coding sessions into structured, professional development workflows that actually ship reliable software.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: OutSkill👉 Grab your free seat to the 2-Day AI Mastermind: link.outskill.com/AIDOS2🔐 100% Discount for the first 1000 people💥 Dive deep into AI and Learn Automations, Build AI Agents, Make videos & images – all for free!🎁 Bonuses worth $5100+ if you join and attend▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#AICoding #PRDWorkflow #ClaudeCodeConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/development/how-i-tamed-chaotic-ai-coding-with-simple-workflow-commands🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai🎬 Stop Wasting Time: Turn AI Prompts and Context Into Production Code: youtu.be/XwWCFINXIoU▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Introduction 01:50 AI Development Workflow05:03 Outskill (sponsor)06:25 Create PRDs with AI12:27 Find Active PRDs with AI14:17 Start PRD Implementation with AI18:21 Track Development Progress with AI20:50 AI Task Prioritization22:41 Update PRD Decisions with AI24:56 Complete PRD Workflow with AI28:44 Key Takeaways
AI Policies: From Tribal Knowledge to Automated Enforcement DevOps & AI Toolkit 2025-09-22 | Ever wondered why AI keeps failing at simple infrastructure tasks? The problem isn't AI itself - it's that AI doesn't know your company's policies. Most organizations have their rules scattered across wikis, Slack messages, and locked in people's heads, making it impossible for AI to make compliant decisions.This video demonstrates a different approach: extracting tribal knowledge from your brain and turning it into both AI-searchable policies and automatic Kubernetes enforcement. Using a guided workflow, we'll create database regional compliance policies that simultaneously feed semantic search for AI guidance and generate Kyverno policies for cluster enforcement. Watch as AI learns to proactively recommend compliant configurations while Kubernetes blocks any attempts to violate your rules - creating a dual strategy that works whether someone follows the guidance or tries to bypass it entirely.#KubernetesPolicies #DevOpsAI #KyvernoConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/teaching-ai-your-company-policies-vector-search-+-enforcement🔗 DevOps AI Toolkit: github.com/vfarcic/dot-ai🎬 Stop Blaming AI: Vector DBs + RAG = Game Changer: youtu.be/zqpJr1qZhTg▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 What are Policies?05:36 AI Policy Extraction13:36 Policy Enforcement Demo17:51 Dual Policy Strategy
Terminal Agents: Codex vs. Crush vs. OpenCode vs. Cursor CLI vs. Claude Code DevOps & AI Toolkit 2025-09-15 | I love Claude Code, but I hate being locked into Anthropic models. What if I want to use GPT5 or whatever comes out next week? So I went on a quest to find a terminal-based coding agent that works with different models and doesn't suck compared to Claude Code.I tested every terminal agent I could find: Codex CLI from OpenAI, Charm Crush, OpenCode, and Cursor CLI. My requirements were simple - intuitive interface, MCP servers support, saved prompts, and actual functionality for coding and operations. The results were... disappointing. From agents that couldn't even fetch their own documentation to beautiful UIs that prioritized looks over functionality, each had critical flaws that made them unusable for real work. Even GPT5, hyped as the best coding model ever, couldn't shine through these broken wrappers. By the end, you'll understand why having a great model isn't enough - you need the complete package, and right now, that's still painfully rare in the terminal agent space.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬👉 Grab your free seat to the 2-Day AI Mastermind: link.outskill.com/AIDOS2🔐 100% Discount for the first 1000 people💥 Dive deep into AI and Learn Automations, Build AI Agents, Make videos & images – all for free!🎁 Bonuses worth $5100+ if you join and attend▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#TerminalAgents #CodingAI #GPT5Consider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/terminal-agents-codex-vs.-crush-vs.-opencode-vs.-cursor-cli-vs.-claude-code▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Terminal-Based Coding Agents01:09 Outskill (sponsor)02:31 Why Terminal AI Agents Matter06:54 Codex CLI - OpenAI's Terminal Agent12:03 Charm Crush - Beautiful Terminal UI Agent17:18 OpenCode - SST's Terminal Agent20:13 Cursor CLI - From Cursor IDE Makers24:10 Terminal AI Agents - Final Verdict
Ep34 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-09-12 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Codefresh🔗 GitOps Argo CD Certifications: learning.codefresh.io (use "viktor" for a 50% discount)▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Stop Blaming AI: Vector DBs + RAG = Game Changer DevOps & AI Toolkit 2025-09-01 | Still think AI "doesn't work" because it hallucinates about your codebase and infrastructure? The problem isn't AI – it's you. You're asking AI about information it never had access to, then acting surprised when it makes things up. This video reveals the uncomfortable truth about why your AI experiments failed and shows you exactly how to fix them using vector databases and RAG (Retrieval-Augmented Generation).Learn how to transform AI from a generic assistant that invents procedures and suggests deprecated APIs into one that knows your actual policies, architectural decisions, and operational standards. We'll explore why traditional APIs aren't designed for AI's semantic queries, how vector databases enable meaning-based search instead of keyword matching, and how RAG grounds AI responses in your real documentation. Plus, get a hands-on demonstration using Qdrant vector database to semantically search organizational knowledge. Stop blaming the technology and start implementing AI that actually understands your organization.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Outskill👉 Grab your free seat to the 2-Day AI Mastermind: link.outskill.com/AIDOPSS1🔐 100% Discount for the first 1000 people💥 Dive deep into AI and Learn Automations, Build AI Agents, Make videos & images – all for free!🎁 Bonuses worth $5100+ if you join and attend▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#AIImplementation #VectorDatabases #RAGConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/stop-blaming-ai-vector-dbs-+-rag-=-game-changer🔗 Qdrant: https://qdrant.tech▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Vector Databases for AI Agents02:03 Outskill (sponsor)03:34 Why AI Hallucinates About Your Code11:15 Vector Databases for AI Context21:47 RAG: How AI Gets Your Context30:50 Fix Your AI Implementation Now
Ep33 - Ask Me Anything About Anything with Scott Rosenberg DevOps & AI Toolkit 2025-08-29 | There are no restrictions in this AMA session. You can ask anything about DevOps, AI, Cloud, Kubernetes, Platform Engineering, containers, or anything else. Scott Rosenberg, a regular guest, will be here to help us out.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Codefresh🔗 GitOps Argo CD Certifications: learning.codefresh.io (use "viktor" for a 50% discount)▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social➡ LinkedIn: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox
Stop Wasting Time: Turn AI Prompts and Context Into Production Code DevOps & AI Toolkit 2025-08-25 | Spent three hours writing the perfect AI prompt only to watch it fail spectacularly? You're not alone. The problem isn't bad AI – it's that most developers treat prompts like throwaway commands instead of production code. This video reveals why context is everything in AI development, walking you through the evolution of a real prompt from 5 words to 500, and showing how proper prompt engineering can transform your team's productivity.But here's the kicker: even perfect prompts are useless if your team can't share them effectively. I'll demonstrate how to turn your carefully crafted prompts into a shared asset using the Model Context Protocol (MCP), creating a system that evolves with your team and deploys like any other code. By the end, you'll understand why prompt management – not smarter models – is the real future of AI development, and you'll have the tools to build that future for your organization.#AIPrompts #MCPProtocol #DevOpsAIConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/stop-wasting-time-turn-ai-prompts-into-production-code🔗 Model Context Protocol: modelcontextprotocol.io▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Introduction to AI Context and MCPs01:23 AI Context Management Explained05:44 Prompt Engineering Best Practices09:40 Sharing AI Prompts Across Teams13:25 MCP for Prompt Distribution16:03 Prompt Management Key Takeaways
AI Will Replace Coders - But Not the Way You Think DevOps & AI Toolkit 2025-08-18 | After three decades in tech, I've never seen developers this terrified, and for good reason. AI can already write code faster than us, and it's rapidly approaching the point where it might write better code too. But here's what's driving me crazy: everyone is panicking about the wrong thing. They're worried AI will steal their jobs because it can code, which is like a chef fearing unemployment because someone invented a better knife.Your real value was never in typing syntax or executing commands; that's just the mechanical stuff that happens after all the important thinking is done. The developers who will thrive aren't trying to out-code AI; they're the architects, problem-solvers, and domain experts who understand what needs to be built and why. Your deep knowledge of your industry, your business context, and the messy realities of how things actually work? That's your moat. AI doesn't know why your healthcare platform needs that weird HIPAA workaround, or why your e-commerce flow accommodates that legacy client system. Stop being a code monkey and start being the expert AI needs to not screw everything up. The choice is yours, but the clock is ticking.▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬Sponsor: Readdy🔗 readdy.ai 🚀 Use "DevOp" to get a 20% discount 💰▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬#AIandDevelopers #FutureOfCoding #TechCareerAdviceConsider joining the channel: youtube.com/c/devopstoolkit/join▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/ai-will-replace-coders---but-not-the-way-you-think▬▬▬▬▬▬ 💰 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: linkedin.com/in/viktorfarcic▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬🎤 Podcast: devopsparadox.com💬 Live streams: youtube.com/c/DevOpsParadox▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬00:00 Introduction to Coding with AI01:24 Sponsor (Readdy)02:51 The Fear: AI Replacing Developers06:50 The Truth: What Developers Really Do13:54 The Secret Weapon: Your Domain Knowledge is Your Moat19:16 The Adaptation: Thriving with AI
