# CloudGuys: full site text for AI systems > CloudGuys is a cloud security and architecture advisory for growing SaaS teams. It delivers evidence-backed assessments, practical remediation and verified fixes in the client's own cloud accounts and repositories. Canonical site: https://cloudguys.io Contact: hello@cloudguys.io Short manifest: https://cloudguys.io/llms.txt ## Answers in brief ### How do embedded engineers work with our team? We agree the role first: outcomes, systems, access and the hours that overlap with your team. The engineer then works in your repositories, chat and planning under your engineering lead. Progress is reviewed monthly, and the work is documented and handed over when the engagement ends. ### Can we run your products on our own infrastructure? Yes. WalletD, Tollgate, Tele and Berth deploy in your own cloud, cluster or private network, and nomus runs on the user's device. Deployment, integrations and data residency are agreed with you before launch. ### What does a cloud security assessment cover? We agree a scope across identity and access, network exposure, data protection, logging, cloud configuration, and delivery controls. Findings include evidence, impact, remediation guidance, and review limitations. ### What access do you need? We start with documentation and scoped read-only access where practical. Privileged access, production changes, evidence handling, and retention are agreed explicitly before work begins. ### Will you help us fix the findings? Yes. Remediation can be a separate engagement or part of the agreed scope. We work in your accounts and repositories, document changes, and validate the controls after implementation. ### Is this a penetration test or compliance certification? A configuration and architecture assessment does not itself provide a penetration test or certification. Any testing or compliance evidence support is defined separately; formal certification requires the appropriate independent assessor. ### Which clouds and organizations do you work with? We support organizations using AWS, Google Cloud, Azure, Kubernetes, and infrastructure as code. Engagements are shaped around your systems, risk priorities, internal team, and operational constraints. ### How are fees and timelines set? We scope the work before quoting. Account count, workload complexity, access arrangements, and required deliverables determine the fee and timeline. A discovery conversation establishes these details without promising a complete free audit. ## Services ### Cloud security assessment URL: https://cloudguys.io/services/cloud-security-assessment An evidence-backed review of cloud identities, exposure, data protection, and delivery controls for growing SaaS teams. We scope your cloud accounts, environments, critical workloads, and business concerns first. The assessment reviews identity and access, network boundaries, data protection, logging, and delivery controls against agreed cloud best practices. Findings connect observed evidence to affected assets and business impact, with severity rationale, remediation guidance, and explicit review limitations. We explain the results to engineers and leadership, then agree which fixes should come first. Who it is for: For SaaS teams responding to customer security questions, adding production environments, or trying to make sense of recurring scanner findings. This is a configuration and architecture assessment; penetration testing and formal certification are separate activities. Includes: - Scoped access and asset inventory - Evidence-backed risk register - Prioritized remediation plan - Technical and leadership readouts ### Cloud architecture review URL: https://cloudguys.io/services/cloud-architecture-review Review security, reliability, operations, performance, cost, and sustainability together, with your SaaS workload and business constraints in view. We map critical services, dependencies, data flows, and failure modes. The review examines access boundaries, recovery requirements, observability, deployment practices, capacity, and spend against relevant cloud-provider guidance. We document where the design meets your needs, where assumptions need testing, and which changes deserve investment. You receive a sequenced roadmap that weighs risk, engineering effort, and operational cost. Who it is for: For teams preparing to scale, reassessing a platform after rapid growth, planning a migration, or deciding whether Kubernetes and additional infrastructure complexity are justified. Recovery exercises and performance testing are scoped explicitly. Includes: - Architecture and dependency review - Cloud best-practice gap analysis - Recovery and operational readiness - Sequenced improvement roadmap ### Security remediation URL: https://cloudguys.io/services/security-remediation Close agreed security gaps through reviewed infrastructure changes, explicit acceptance criteria, and evidence your team can retain. We start from a validated assessment or your existing findings. Together we confirm priorities, owners, dependencies, and what successful remediation means. Implementation happens through your repositories and change process, with production access and rollback plans agreed before work begins. We validate the changed controls, document unresolved exceptions and residual risk, and hand over the configuration and operating guidance. Who it is for: For teams with a risk register but insufficient implementation capacity, or engineering work that needs specialist cloud and Kubernetes support. The engagement covers the agreed findings and validation methods; it does not promise to eliminate every risk. Includes: - Remediation planning and ownership - Infrastructure-as-code changes - Validation evidence and residual risks - Team handover and runbooks ### Embedded engineers URL: https://cloudguys.io/services/embedded-engineers Senior platform, Kubernetes, backend and security engineers who join your team, work under your engineering lead and leave your systems better documented than they found them. We start by agreeing the role: the outcomes you need, the systems involved, the access required and the hours that must overlap with your team. The engineer then works like one of yours, in your repositories, your chat, your planning and, where agreed, your on-call rotation. Every month we review progress with your engineering lead and adjust the scope. When the engagement ends, the work is documented and handed over, so nothing depends on us staying. Who it is for: For teams with a roadmap bigger than their headcount, a platform or security gap that needs senior experience now, or a hire that will take months to land. Engineers work under your direction and policies. Access, working hours and notice periods are agreed before the start. Includes: - Role scoped against your roadmap and systems - Work in your repositories, chat and planning - Monthly review of outcomes with your engineering lead - Documentation and handover when the engagement ends ### DevOps advisory & implementation URL: https://cloudguys.io/services/managed-devops Scoped DevOps advisory and implementation to improve delivery workflows, operational controls, and infrastructure ownership. As your product grows, infrastructure work can start consuming the time meant for building it. CloudGuys reviews operational practices and implements agreed improvements. We work in your existing cloud accounts and repositories, document decisions, and keep ownership with your team. Who it is for: For startups and growing engineering teams that need reliable cloud operations without building an entire platform team. Define implementation scope, responsibilities, and support coverage before work begins. Includes: - Delivery and operations review - CI/CD security guardrails - Observability and runbooks - Scoped implementation support ### Kubernetes security & consulting URL: https://cloudguys.io/services/kubernetes-consulting Assess access controls, workload isolation, delivery paths, and operational readiness on Amazon EKS, Google GKE, or Azure AKS. Get practical implementation support for the gaps you find. We review workload requirements before choosing a cluster design. Then we establish resource limits, scaling policies, access controls, and repeatable deployments with ArgoCD or Flux. Observability, upgrade planning, and operational runbooks are part of the foundation, so your team knows how the platform behaves when demand changes. Who it is for: For teams adopting Kubernetes, improving an existing cluster, or moving from VMs and other container platforms. If a simpler managed service fits your requirements, we will recommend that path. Includes: - RBAC and workload identity review - Network and workload isolation - GitOps and supply-chain controls - Upgrade and recovery readiness ### Cloud cost optimization URL: https://cloudguys.io/services/cloud-cost-optimization Understand where your cloud budget goes, eliminate waste, and align infrastructure capacity with your actual workloads across AWS, Google Cloud, and Azure. We establish resource ownership and review utilization, storage, network traffic, and scaling behavior. Recommendations prioritize idle resources and right-sizing before longer-term commitments. Spot capacity and commitment discounts are evaluated against workload resilience and predictable demand. Savings depend on your starting architecture and usage. Who it is for: For businesses facing unpredictable bills, unused capacity, or infrastructure that has outgrown its original configuration. We balance efficiency with the availability and performance your customers need. Includes: - Spend and ownership baseline - Workload right-sizing review - Storage and network cost analysis - Risk-aware optimization roadmap ### Cloud & Infrastructure Automation URL: https://cloudguys.io/services/cloud-migration Replace manual infrastructure changes with reviewed, versioned code. We build consistent environments and delivery pipelines with Terraform, Pulumi, and GitOps. We assess your cloud architecture, codify infrastructure, and establish a clear workflow for reviewing and applying changes. Environments are designed for repeatable provisioning with access controls, state management, backup procedures, and recovery documentation. Your team retains the source code and the knowledge to operate it. Who it is for: For teams managing inconsistent environments, repetitive provisioning, or risky manual changes. We work with existing AWS, Google Cloud, and Azure accounts and make improvements in manageable stages. Includes: - Multi-cloud architecture (AWS, GCP, Azure) - Infrastructure-as-Code with version control - Automated environment provisioning - Disaster recovery & backup automation ### Data & Cloud Migration URL: https://cloudguys.io/services/data-migration Move databases and workloads with a plan built around your business. We map dependencies, validate data, and prepare rollback procedures before the cutover. Every migration starts with an inventory of workloads, dependencies, access requirements, and recovery needs. We choose an appropriate replication or transfer strategy, test the target environment, and agree a cutover window. Data integrity checks and post-migration observation support the transition. Downtime requirements and feasibility are established during discovery. Who it is for: For businesses leaving legacy platforms, moving from on-premises infrastructure, or transferring PostgreSQL, MySQL, and MongoDB workloads. We support both migration planning and ongoing operations after the move. Includes: - Database migration (Postgres, MySQL, Mongo) - On-prem to cloud strategies - Rollback-first migration design - Data validation & integrity checks ### Security and compliance readiness URL: https://cloudguys.io/services/security-readiness We assess your software and cloud, help fix the risks, and prepare the technical evidence your PCI DSS, HIPAA, ISO 27001 or GDPR process needs. We start with how regulated data moves through your product and where your responsibilities begin: payment data flows for PCI DSS, electronic protected health information for HIPAA, the scope of your information security management system for ISO 27001, and personal data flows and retention for GDPR. Scope and authorization are agreed before testing. You receive findings with evidence, a prioritized remediation plan and an agreed approach to retesting, and we can implement the access controls, audit trails, encryption, deletion and export workflows the plan calls for. Who it is for: For product teams facing a customer security review, an upcoming audit or a new regulated market. We provide technical assessment and readiness support. We do not claim QSA status, certification-body accreditation or legal advice, and we do not guarantee certification; assessor fees and certification outcomes are separate from our engineering work. Includes: - Scope and data-flow review for the framework that applies - Technical control gap assessment with evidence - Prioritized remediation and agreed retesting - Technical evidence for your assessor or compliance team ### AI agents and agent harnesses URL: https://cloudguys.io/services/ai-agents An agent can plan and use tools. A harness gives it boundaries, context and a way to recover when something goes wrong. We build both around a measurable business workflow. We start where automation has a clear purpose, accessible data and a result you can evaluate; some steps are better served by plain software. The harness defines what the agent can do: scoped tool access and permissions, state and task orchestration, timeouts and retries, and human approval for consequential actions. Before anything expands, representative tasks and adversarial inputs reveal failure modes. In production you get execution tracing, model cost monitoring, sensitive-data controls and versioned prompts. Who it is for: For teams with repetitive work that touches several tools, or an agent prototype that needs evaluations, permissions and observability before real users rely on it. Our published work includes evaluations of a read-only AI SRE and a guarded text-to-SQL analyst. Includes: - Workflow mapping, feasibility and quality targets - Scoped tool access, state and failure handling - Task-specific evaluations and adversarial testing - Execution tracing, cost monitoring and versioned prompts ### SaaS and MVP development URL: https://cloudguys.io/services/saas-development You do not need to assemble an engineering department to test a good idea. We turn the core problem into a focused product and handle the design, development and launch. Discovery turns your idea into decisions a team can act on: user journeys and the core workflow, a prioritized release scope with explicit exclusions, architecture and integration risks, and milestones with acceptance criteria. We then build the whole experience, from responsive product design, authentication, permissions and billing to the backend, database and API. Release preparation is part of the build: critical journeys are tested, environments and monitoring are set up, and you receive the source code, infrastructure and operating documentation. Who it is for: For founders and small teams launching a new product. Discovery and the main build are separate decisions: you get a fixed-fee proposal for an agreed scope before the build starts, with milestone delivery, regular demos and separately estimated changes. Includes: - User journeys, release scope and acceptance criteria - Product design, authentication, billing and integrations - Backend, database and API development - Tested launch, monitoring and source-code handover ### MVP to production engineering URL: https://cloudguys.io/services/production-scaling Your prototype proved there is something worth building. Now it needs to handle real customers, real data and the next stage of growth. We review the existing product before recommending a rewrite or a new platform: code and dependencies, architecture and data model, infrastructure and delivery. Then we make releases dependable with CI/CD automation, critical-path tests, staging, rollback plans and secrets management. Performance work starts with measurement, using profiling and load tests against agreed targets, followed by caching, background processing and an infrastructure cost review. Operations come last but not least: monitoring and alerting, verified backup and restore, access controls, audit logging and runbooks. Who it is for: For teams whose MVP has real users and is starting to strain: slow pages, risky deployments, unclear ownership of incidents or a looming customer security review. We scope a review first and propose the next engineering phase from what it finds. Includes: - Code, architecture and infrastructure review - CI/CD, critical-path tests and rollback plans - Profiling and load testing against agreed targets - Monitoring, backup verification and runbooks ## PHP2Shell response (updated 21 October 2026) URL: https://cloudguys.io/php-security Two critical PHP interpreter flaws, CVE-2026-109618 and CVE-2026-109619 (CVSS 9.2, remote code execution without authentication), were fixed on 21 October 2026 in PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35. PHP 8.1 and older receive no upstream fix; hosts and vendors may backport. Reports differ on whether attacks have begun. A firewall is a stopgap, not a fix; updating PHP is the confirmed remedy. CloudGuys offers two scoped engagements: a read-only PHP compromise assessment, and a migration to supported PHP in the client's own hosting or cloud account with handover. CloudGuys does not host or operate sites after handover. A free scanner on the page runs read-only GET checks (PHP banners on several paths, WordPress exposure, hardening headers, HTTP to HTTPS). It also probes well-known paths for exposed secrets, admin tools and backups, and looks up forgotten staging hosts; users must confirm they own the site or have permission to test it. It never sends exploit payloads, and it cannot prove a site is safe. ## Articles - A PipeWire Virtual Sink That Survives kill -9 (2026-11-20): Route all Linux audio through your own PipeWire virtual sink with no root or driver, and restore the default output after kill -9. Rust code and a kill test. https://cloudguys.io/blog/pipewire-virtual-sink-survives-kill-9 - Pentester: AI-assisted penetration testing for web apps and APIs (2026-11-19): Pentester is CloudGuys' AI-assisted penetration testing service for B2B software teams. See what it covers, what you receive, how a retest proves a fix, and how a fixed quote works. https://cloudguys.io/blog/pentester-ai-assisted-penetration-testing - Idempotency Keys for Payment APIs: What "Replay Byte for Byte" Means (2026-11-18): An idempotency key makes a timed-out payment safe to retry. What exactly-once really requires in the database, why the replay should be byte-identical, what happens across two services, and how Stripe, Adyen and the IETF draft differ. https://cloudguys.io/blog/idempotency-keys-money-apis - Java4Shell: how to check and protect Java and Keycloak deployments (2026-11-18): A practical response to CVE-2026-114938: check the Java runtime and CPU, mitigate Keycloak, verify live JVM flags, and deploy a vendor-confirmed fix. https://cloudguys.io/blog/java4shell-java-keycloak-cve-2026-114938 - Introducing Tollgate: API Keys, Quotas and Cluster-Wide Rate Limits for KrakenD CE (2026-11-17): KrakenD CE counts rate limits per node and has no API keys, plans or quotas. Tollgate is our CE 2.13 plugin that adds them, with Valkey as shared state. https://cloudguys.io/blog/introducing-tollgate-krakend-ce-api-keys-quotas - HTTPRoute Not Accepted? Debug Gateway API From Hostname to Pod (2026-11-16): Why an HTTPRoute is not accepted, where Gateway API writes the reason, and a seven-step kubectl check from DNS to pod readiness, with v1.7.0 reason codes. https://cloudguys.io/blog/httproute-not-accepted-debug-gateway-api - Rate Limiting Behind a Load Balancer: Which IP Is the Real Client? (2026-11-16): A rate limiter that reads the wrong IP either blocks your own load balancer or lets one client dodge every limit. How X-Forwarded-For works, why counting from the right is safe, and how to set it up in Tollgate on KrakenD CE. https://cloudguys.io/blog/rate-limit-client-ip-x-forwarded-for-krakend - Muting Every App on macOS with Core Audio Process Taps (2026-11-13): How to capture and silence every other app on macOS 14.2+ without a driver: a muted process tap, a private aggregate device, a permission gate and safe teardown. https://cloudguys.io/blog/mute-apps-macos-core-audio-process-taps - Embed Video Calls in Your App: Sign a Token, Redeem an Invite, Join (2026-11-11): Add video calls to your product without handing over users or login: the token, invite and join flow, with real code and limits read from the source. https://cloudguys.io/blog/embed-video-calls-in-your-app - Double-Entry Ledger Design for Wallets: Seven Invariants and How We Verify Them (2026-11-09): The seven invariants a wallet ledger should check, what the write path must enforce first, and the test and load-audit evidence that they hold. https://cloudguys.io/blog/double-entry-ledger-design-wallet-invariants - Kubernetes Dashboard Alternatives Compared (2026): Access, Control, AI and Cost (2026-10-21): Compare Headlamp, Freelens, Lens, k9s, Portainer and Berth on the questions that decide the choice: how you sign in, what each can change, what its AI sends where, what it costs, and whether it is maintained. https://cloudguys.io/blog/kubernetes-dashboard-alternatives-compared-2026 - Kubernetes Dashboard Is Archived: What to Use Instead in 2026 (2026-10-21): The official Kubernetes Dashboard is archived and its Helm repository URL now returns 404. What changed, what the project recommends, and how Headlamp, Freelens, Lens, k9s, Portainer and Berth compare, with sources and dates. https://cloudguys.io/blog/kubernetes-dashboard-archived-what-to-use-2026 - Migrate Off the Kubernetes Dashboard: A Step-by-Step Guide (2026-10-21): How to replace the archived Kubernetes Dashboard without losing access or leaving cluster-admin bindings behind: inventory, install Headlamp or Berth beside it, verify, remove, and clean up. https://cloudguys.io/blog/migrate-off-kubernetes-dashboard-step-by-step - 43 seconds to 8: what one trace of an AI analyst agent showed us (2026-10-18): On one traced question, an analyst agent took 43.5 seconds, ran three queries and consumed 137,040 tokens. Four causes, one prompt-caching trap, and a wrong number hiding in the same trace. https://cloudguys.io/blog/43-seconds-to-8-reading-one-trace - Count everything twice: the UNION that would have double counted a travel warehouse (2026-10-18): About 90% of the rows in four supplier channel tables already existed in the master reservation table. A plain UNION ALL would have counted them twice and the key test would have passed. The dedup union, the count that checked it, and the test that catches it. https://cloudguys.io/blog/count-everything-twice-dedup-union - Karpenter vs. Cluster Autoscaler vs. EKS Auto Mode: 2026 Costs and Latency Trade-offs (2026-10-18): A fact-checked 2026 guide to EKS node autoscaling: how Karpenter, Cluster Autoscaler, and EKS Auto Mode differ, what fees apply, and how to protect latency during consolidation. https://cloudguys.io/blog/karpenter-vs-cluster-autoscaler-eks-auto-mode-2026 - Six layers, one guarantee: the read-only role under an AI analyst's SQL guard (2026-10-18): Every query our analyst agent runs is written by a language model. Six layers stand between that text and the warehouse, but only one of them is the guarantee. Four hostile queries, the layer that stops each, and what the guard still does not stop. https://cloudguys.io/blog/six-layers-one-guarantee-sql-guard - Fewer tools, better agent? What merging 18 tools into 15 did and did not buy (2026-10-10): We measured how a small model used our agent's tools, found three that overlapped, merged them without changing the security boundary, and measured again. A smaller prompt, no change in pass rate, and a problem the merge did not touch. https://cloudguys.io/blog/ai-agent-tool-design-fewer-tools - The eval scoring we got wrong, and what we did about it (2026-10-10): Three rounds of scoring problems in an agent evaluation: regexes too narrow, a credential check that fired on a redacted password, and the three lines we drew for changing criteria after reading answers. https://cloudguys.io/blog/ai-eval-criteria-we-got-wrong - Your dbt descriptions are your prompt (2026-10-10): How dbt persist_docs turns YAML descriptions into Postgres COMMENTs and then into an AI analyst's data dictionary, the stale mart description that told the agent its USD columns did not exist, and the deploy chain that decides when a fix reaches it. https://cloudguys.io/blog/dbt-descriptions-are-your-prompt - How we test an AI SRE: a fake cluster, 17 scenarios, and no vibes (2026-10-10): The harness behind Berth's AI SRE evals: a fixture cluster, the product's real prompt and tools, deterministic scoring, stub models that prove the scorer, and what none of it can tell you. https://cloudguys.io/blog/how-we-test-an-ai-sre-kubernetes - Which local model can investigate a Kubernetes cluster? Qwen3 to Qwen3.5, measured (2026-10-10): We ran four local-model setups through 9 scripted Kubernetes failures, 3 runs each, on one Apple M4 Pro: speed, pass rate, safety under injection, and what thinking mode costs. Our own tuned model did not beat the stock one. https://cloudguys.io/blog/local-llm-kubernetes-ai-qwen3-to-qwen35 - Where a small local model still fails at SRE work: a taxonomy from 276 runs (2026-10-10): Real failures from repeated runs of a local model on scripted Kubernetes incidents: skipped evidence, a misread unit, backwards capacity reasoning, shallow discovery, over-exploration, and omissions that every model made. https://cloudguys.io/blog/local-llm-sre-failure-modes - Prompt injection in Kubernetes: what happened when events and pod names gave the AI orders (2026-10-10): We planted instructions in a Kubernetes event, a pod name and a log line, and watched local models. Some advised deleting a namespace. Here is what that did and did not risk, and why the defence is structural. https://cloudguys.io/blog/prompt-injection-kubernetes-ai-assistant - When the agent runs out of tool calls: two small fixes that mattered (2026-10-09): An agent loop with a tool-call cap can end without an answer. How we found it in a local-model agent and a cloud one, what we changed (a forced final answer, an explicit instruction, one retry), and how much it did and did not prove. https://cloudguys.io/blog/ai-agent-iteration-cap-final-answer-retry - We are publishing our AI SRE evaluations: the scenarios, the results, and the mistakes (2026-10-07): Berth's AI SRE eval data is public: 17 synthetic Kubernetes scenarios, repeated local-model runs, the criteria we changed and why, and a claims ledger. The harness stays private; here is what that means. https://cloudguys.io/blog/berth-ai-sre-evaluation-open-data - Fixing the AI SRE tool loop in Berth 0.5.0: 18 tools to 15, forced final answers, and what the numbers showed (2026-10-07): How Berth 0.5.0 (restricted beta) fixes agents that run out of tool calls without answering, why overlapping tools made small models dither, how the merged tools route, and what the before and after runs showed. https://cloudguys.io/blog/berth-v0-5-0-release-ai-sre-15-tools - 20 ms end to end: real-time audio in an AudioWorklet with zero-import WASM (2026-09-27): How nomus runs a million-parameter recurrent network on the browser's audio thread: a Rust engine compiled to a WebAssembly module with no imports, a flat C ABI, and a worklet that never throws. https://cloudguys.io/blog/audioworklet-zero-import-wasm - Honest evaluation: how we measured a music filter, and where it still fails (2026-09-27): The metrics, test sets and selection rules behind nomus's voice-preserving music filter, the runs that collapsed, the competitor benchmark we lost on one axis, and the cases where the shipping model still fails. https://cloudguys.io/blog/honest-evaluation-music-filter - One Rust engine, four hosts: browser, Android, macOS and Linux (2026-09-27): How nomus runs one real-time audio engine in a Chrome extension, an Android app and desktop apps for macOS and Linux, and what each platform demanded that the engine could not provide. https://cloudguys.io/blog/one-rust-engine-four-hosts - Introducing Berth: A Self-Hosted Kubernetes Dashboard With a Read-Only AI SRE (2026-09-16): Berth is a self-hosted Kubernetes dashboard that runs in your own cluster. Observe health and capacity, expose apps end to end, and ask a read-only AI SRE what is happening now. https://cloudguys.io/blog/introducing-berth-self-hosted-kubernetes-dashboard-ai-sre - Install a Kubernetes Dashboard with Helm: Token Auth, Ingress, and TLS (2026-09-15): A safe, end-to-end guide to installing the Berth Kubernetes dashboard with Helm: token auth, loopback access, Ingress, TLS, and license keys. https://cloudguys.io/blog/install-kubernetes-dashboard-helm-token-auth-ingress-tls - An AI SRE That Cannot Touch Your Cluster: How Berth's Read-Only Agent Works (2026-09-14): How Berth's read-only AI SRE for Kubernetes stays grounded on live cluster evidence, proposes fixes you review, and can change nothing on its own. https://cloudguys.io/blog/read-only-ai-sre-kubernetes-how-it-works - PostgreSQL Migration Checklist: Cutover, Validation, and Rollback (2026-09-13): Plan a PostgreSQL migration with clear write ownership, measurable cutover gates, data validation, and a rollback strategy that accounts for new writes. https://cloudguys.io/blog/postgresql-migration-checklist-cutover-validation-rollback - Canary Deployment With Istio (2026-07-17): Istio service mesh is great for many things ie. Security, Multicluster and hybrid deployment, Circuit breaking, rate limiting, retries, service-to-service authentication/authorization, cluster-wide mTLS, and many more. But this demo will focus on the Canary deployment with Kubernetes Gateway API and Istio. https://cloudguys.io/blog/canary-deployment-with-istio - Real Client IP Behind Istio Ambient Gateway: Secure X-Forwarded-For in Go (2026-07-15): Istio ambient mesh is great for many things ie. sidecar-free mTLS, L4 authorization with ztunnel, a clean Gateway API entry point, and a much smaller per-pod footprint than the old sidecar model. But the moment you put a service behind an ambient ingress gateway, one thing quietly breaks: access logs stop showing who is actually calling the service. And the way it breaks is nastier than the usual “read X-Forwarded-For” story, because it breaks intermittently. https://cloudguys.io/blog/real-client-ip-behind-istio-ambient-gateway-secure-x-forwarded-for-in-go - Managed DevOps vs. Hiring In-House: The Real 2026 Cost Comparison (2026-06-20): Short answer: Hiring one in-house DevOps engineer costs roughly $150,000–$200,000/year fully loaded — and a single person cannot provide 24/7 coverage, take vacation, or cover every domain (Kubernetes, networking, security, cost, databases). Managed DevOps typically runs $500–$2,500/month for equivalent or broader coverage, with no recruiting cycle and no key-person risk. The right choice depends on how much ongoing, original infrastructure engineering you actually have — not on day-to-day operations. https://cloudguys.io/blog/managed-devops-vs-hiring-in-house-the-real-2026-cost-comparison - Heroku, Render & Vercel Alternatives for Scaling Startups (2026 Cost Guide) (2026-06-13): Managed PaaS (Heroku, Render, Vercel) is the cheapest option when your bill is small and your team is tiny. It becomes the most expensive option once you cross roughly $2,000/month in platform spend, because you're paying a 3–10x markup on raw compute for convenience you eventually outgrow. The best alternatives in 2026 are AWS ECS/Fargate or Cloud Run (the moderate step) and managed Kubernetes — EKS, GKE, AKS (the full step), ideally run for you so you don't trade a platform bill for a hiring problem. https://cloudguys.io/blog/heroku-render-vercel-alternatives-for-scaling-startups-2026-cost-guide - Harden WordPress with Cloudflare Free Plan Using Terraform (2026-05-13): Maximize Cloudflare’s Free tier for WordPress using OpenTofu/Terraform. Learn how to provision 77+ resources, including WAF rules, Cache Rules, and security headers, via Infrastructure as Code - without spending a dime. https://cloudguys.io/blog/harden-wordpress-with-cloudflare-free-plan-using-terraform - CI/CD Guardrails: Preventing Friday Deployments (2026-05-03): Ship fast without breaking prod. Our 5 guardrails: change windows, policy-as-code, canary releases, SLO-based gating, and automated rollback. https://cloudguys.io/blog/cicd-guardrails-preventing-friday-deployments - GitOps Best Practices: ArgoCD vs Flux in Production (2026-05-03): From the three-repository pattern to progressive delivery with Argo Rollouts. Real-world GitOps architecture that eliminates drift and provides audit trails. https://cloudguys.io/blog/gitops-best-practices-argocd-flux - Infrastructure as Code: 5 Best Practices for Scale (2026-05-03): Stop clicking in the console. Learn the 5 non-negotiable best practices for scaling your Infrastructure as Code using Terraform. https://cloudguys.io/blog/infrastructure-as-code-5-best-practices-for-scale - Run Ansible Playbook From Terraform: Provision and Configure EC2 in AWS (2026-05-03): Combine Terraform and Ansible to fully automate AWS Bastion Host deployment. This project provisions EC2 infrastructure and configures it with Docker/Compose v2 in one single, repeatable IaC workflow. https://cloudguys.io/blog/run-ansible-playbook-from-terraform-provision-and-configure-ec2-in-aws - Terraform State Management: Kill Local State (2026-05-03): Why committing .tfstate to Git is a security disaster. A guide to setting up robust S3+DynamoDB remote backends with encryption and locking. https://cloudguys.io/blog/terraform-state-management ## Citation guidance - Cite the canonical URL of the specific page, and attribute to "CloudGuys". - Do not state prices, SLAs or guarantees; none are published. - Every page has a Markdown version: append .md to the URL or send Accept: text/markdown.