<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Nubyron · Engineering notebook</title><description>Practical notes on cloud architecture, DevOps, Kubernetes, reliability, security and FinOps.</description><link>https://nubyron.com/</link><language>en</language><item><title>82% will increase their digital budget: spending more is not enough</title><link>https://nubyron.com/blog/spanish-companies-increase-digital-transformation-budget-ai/</link><guid isPermaLink="true">https://nubyron.com/blog/spanish-companies-increase-digital-transformation-budget-ai/</guid><description>Digital investment is rising and AI is becoming a priority. The challenge is no longer approving more budget, but turning it into a secure, operable platform tied to results.</description><pubDate>Sat, 22 Aug 2026 06:00:00 GMT</pubDate><category>Digital transformation</category><category>Artificial intelligence</category><category>Cloud</category><category>Cybersecurity</category><category>Technology strategy</category></item><item><title>What we&apos;d review in the first 48 hours of inheriting an Azure environment</title><link>https://nubyron.com/blog/azure-infrastructure-assessment-first-48-hours/</link><guid isPermaLink="true">https://nubyron.com/blog/azure-infrastructure-assessment-first-48-hours/</guid><description>A senior Azure assessment does not start by replacing tools. It starts by identifying what sustains production, where risk sits and what can recover.</description><pubDate>Fri, 21 Aug 2026 09:00:00 GMT</pubDate><category>Azure</category><category>Cloud Assessment</category><category>DevOps</category><category>Security</category><category>FinOps</category></item><item><title>The Day Someone Drops a Table in Production, You&apos;ll Find Out if You Truly Have Backups</title><link>https://nubyron.com/blog/database-backups-recovery-mysql-pitr/</link><guid isPermaLink="true">https://nubyron.com/blog/database-backups-recovery-mysql-pitr/</guid><description>An accidental DELETE, DROP, or TRUNCATE can escalate into a critical incident. Backups, MySQL binary logs, PITR, and restore drills make all the difference.</description><pubDate>Tue, 18 Aug 2026 10:00:00 GMT</pubDate><category>Backups</category><category>Databases</category><category>MySQL</category><category>Point-in-Time Recovery</category><category>PITR</category><category>Disaster Recovery</category><category>SRE</category><category>Cloud Maintenance</category></item><item><title>Eliminating Kubernetes alert fatigue: from noisy metrics to actionable SLOs</title><link>https://nubyron.com/blog/kubernetes-alert-fatigue-slo-actionable-alerts/</link><guid isPermaLink="true">https://nubyron.com/blog/kubernetes-alert-fatigue-slo-actionable-alerts/</guid><description>Why threshold-based alerts create noise, wake up on-call engineers for harmless spikes, and how to implement multi-burn-rate SLO alerting in Prometheus.</description><pubDate>Sun, 16 Aug 2026 15:00:00 GMT</pubDate><category>Kubernetes</category><category>SRE</category><category>Observability</category><category>Prometheus</category><category>SLO</category><category>Alertmanager</category></item><item><title>KubeBolt on OpenShift 4.20: when a 25-second timeout was hiding 138 MiB of Events</title><link>https://nubyron.com/blog/kubebolt-openshift-timeout-events-informer/</link><guid isPermaLink="true">https://nubyron.com/blog/kubebolt-openshift-timeout-events-informer/</guid><description>KubeBolt reached Kubernetes, authenticated and queried resources, yet declared the cluster unreachable. An A/B test isolated 138 MiB of Events as the outlier.</description><pubDate>Sun, 16 Aug 2026 12:00:00 GMT</pubDate><category>Kubernetes</category><category>OpenShift</category><category>KubeBolt</category><category>Troubleshooting</category><category>Platform Engineering</category></item><item><title>Cloud Cost Optimization: A Practical Guide for Engineering Teams</title><link>https://nubyron.com/blog/cloud-cost-optimization-practical-guide/</link><guid isPermaLink="true">https://nubyron.com/blog/cloud-cost-optimization-practical-guide/</guid><description>A practical guide to connecting engineering, finance and business through cloud cost visibility, ownership and economic decisions.</description><pubDate>Sun, 16 Aug 2026 09:00:00 GMT</pubDate><category>FinOps</category><category>Cloud Cost Optimization</category><category>Cloud Governance</category></item><item><title>How to choose a cloud migration partner</title><link>https://nubyron.com/blog/how-to-choose-cloud-migration-partner/</link><guid isPermaLink="true">https://nubyron.com/blog/how-to-choose-cloud-migration-partner/</guid><description>A practical framework for comparing cloud migration companies before architecture, cost and production risk become contractual surprises.</description><pubDate>Sat, 15 Aug 2026 13:00:00 GMT</pubDate><category>Cloud Migration</category><category>Migration Partner</category><category>Cloud Architecture</category><category>Rollback</category><category>Vendor Selection</category></item><item><title>We are building software faster than ever. But who will operate it?</title><link>https://nubyron.com/blog/ai-software-development-operations/</link><guid isPermaLink="true">https://nubyron.com/blog/ai-software-development-operations/</guid><description>AI is reducing the friction of building software. Understanding, securing and sustaining it in production does not disappear; the work moves elsewhere.</description><pubDate>Thu, 13 Aug 2026 09:00:00 GMT</pubDate><category>AI software development</category><category>Platform Engineering</category><category>Cloud Operations</category><category>SRE</category><category>Observability</category></item><item><title>If your cloud cost controls only alert you, they are already too late</title><link>https://nubyron.com/blog/ai-finops-cloud-cost-controls/</link><guid isPermaLink="true">https://nubyron.com/blog/ai-finops-cloud-cost-controls/</guid><description>AI workloads can increase spend faster than traditional FinOps processes can respond. The next step is not only earlier detection, but deciding when to intervene without putting the business at risk.</description><pubDate>Tue, 11 Aug 2026 09:00:00 GMT</pubDate><category>AI FinOps</category><category>Cloud costs</category><category>Platform Engineering</category><category>SRE</category><category>Artificial intelligence</category></item><item><title>DevOps as a Service explained: how the model works and when it fits</title><link>https://nubyron.com/blog/devops-as-a-service/</link><guid isPermaLink="true">https://nubyron.com/blog/devops-as-a-service/</guid><description>Learn how the DevOps as a Service operating model works, what responsibilities it can cover, and how it compares with internal hiring and focused consulting.</description><pubDate>Sat, 08 Aug 2026 14:07:00 GMT</pubDate><category>DevOps</category></item><item><title>Kubernetes shouldn&apos;t give you mini-heart attacks</title><link>https://nubyron.com/blog/kubernetes-should-not-give-you-mini-heart-attacks/</link><guid isPermaLink="true">https://nubyron.com/blog/kubernetes-should-not-give-you-mini-heart-attacks/</guid><description>Knowing how to deploy a cluster is not the same as knowing how to operate it. Kubernetes brings scalability, but also a considerable operational surface. Discover how to operate it with confidence.</description><pubDate>Sat, 01 Aug 2026 19:36:33 GMT</pubDate><category>Kubernetes</category></item><item><title>Why staging works and production fails</title><link>https://nubyron.com/blog/why-staging-works-production-fails/</link><guid isPermaLink="true">https://nubyron.com/blog/why-staging-works-production-fails/</guid><description>The same commit behaves differently when data, permissions, traffic and dependencies change. A diagnostic for teams using staging as a lucky charm.</description><pubDate>Fri, 10 Jul 2026 09:00:00 GMT</pubDate><category>Staging</category><category>Production</category><category>Deployments</category><category>Parity</category><category>Agencies</category></item><item><title>Multi-client infrastructure for AI agencies: what to standardise and what to isolate</title><link>https://nubyron.com/blog/multi-client-infrastructure-ai-agencies/</link><guid isPermaLink="true">https://nubyron.com/blog/multi-client-infrastructure-ai-agencies/</guid><description>A practical guide to delivering agents, RAG systems and automations across clients without multiplying environments, access and support.</description><pubDate>Fri, 03 Jul 2026 09:00:00 GMT</pubDate><category>AI agencies</category><category>Multi-client infrastructure</category><category>CI/CD</category><category>Platform Engineering</category></item><item><title>7 signs infrastructure is slowing down your automation agency</title><link>https://nubyron.com/blog/signs-infrastructure-slowing-automation-agency/</link><guid isPermaLink="true">https://nubyron.com/blog/signs-infrastructure-slowing-automation-agency/</guid><description>The problem rarely starts with a major outage. It starts with waiting, unclear access, manual work and clients that no longer produce the same margin.</description><pubDate>Fri, 26 Jun 2026 09:00:00 GMT</pubDate><category>Automation agencies</category><category>Infrastructure</category><category>Scalability</category><category>Operations</category></item><item><title>European compliance for AI agencies: 12 questions before deployment</title><link>https://nubyron.com/blog/european-compliance-infrastructure-ai-agencies/</link><guid isPermaLink="true">https://nubyron.com/blog/european-compliance-infrastructure-ai-agencies/</guid><description>Hosting data in Europe helps, but it does not resolve contracts, subprocessors, access, traceability or responsibility for an AI system on its own.</description><pubDate>Fri, 19 Jun 2026 09:00:00 GMT</pubDate><category>GDPR</category><category>AI Act</category><category>AI agencies</category><category>Data sovereignty</category></item><item><title>How to reduce the DevOps bus factor in a software agency</title><link>https://nubyron.com/blog/reduce-devops-bus-factor-agency/</link><guid isPermaLink="true">https://nubyron.com/blog/reduce-devops-bus-factor-agency/</guid><description>How to find and reduce dependency on the one person controlling Terraform, CI/CD, access and production.</description><pubDate>Fri, 05 Jun 2026 09:00:00 GMT</pubDate><category>Bus factor</category><category>DevOps for agencies</category><category>Continuity</category><category>Terraform</category></item><item><title>Reference architecture: PDF processing on Azure</title><link>https://nubyron.com/blog/from-diagram-to-cloud-infrastructure/</link><guid isPermaLink="true">https://nubyron.com/blog/from-diagram-to-cloud-infrastructure/</guid><description>How to simplify a PDF processing design and turn it into an event-driven, modular pipeline defined with Terraform.</description><pubDate>Fri, 02 Jan 2026 15:08:03 GMT</pubDate><category>Azure</category><category>Terraform</category><category>Document Intelligence</category><category>Event-driven</category></item></channel></rss>