Blog

Articles on cloud architecture, structured AI development, and lessons from enterprise projects.

May 27, 2026

Tracing AI Agents Is Not Enough

Traces tell you what an agent did, not what it was allowed to do. If you treat observability as governance, you’ll keep shipping incidents—because the control plane (identity, policy, execution-time enforcement) still isn’t there.

#ai-agents #observability #policy #platform-engineering #opentelemetry #jaeger

May 26, 2026

Policy Must Move Left of Kubernetes

Admission control catches bad manifests, but it’s a costly last resort when the decision was already made in templates, charts, or PRs. Move policy into the platform contract so safe defaults and workflows prevent misconfigurations before runtime.

#platform-engineering #kubernetes #policy-as-code #devops #opa #backstage

May 25, 2026

OpenTelemetry Is Becoming the Default Observability Layer

OpenTelemetry’s graduation turns observability from a backend shopping list into platform infrastructure. The real fight is who controls collection, routing, and sampling—because OTLP support won’t save you from lock-in or runaway telemetry costs.

#observability #opentelemetry #platform-engineering #devops #kubernetes #governance

May 21, 2026 three-ways-ai

AI Coding Agents Need Policy, Not Hope

Coding agents already run with real user permissions—so a bad prompt or model slip becomes an operational incident, not a harmless mistake. Prempti-style policy interception can stop tool calls before damage, but you still need visibility and scoped permissions.

#ai-coding-agents #policy-as-code #security-governance #falco #devops

May 21, 2026

Kubernetes Is Becoming the AI Runtime

Kubernetes can run AI workloads, but the real shift is organizational: platform teams may have to own AI execution like they already own everything else. NetEase’s Tmax shows the bottleneck isn’t scheduling—it’s data movement and cold-start latency.

#kubernetes #platform-engineering #devops #ai-infrastructure #data-movement #gpu-scheduling

Apr 16, 2026 three-ways-ai

Your AI-Native DevOps Playbook

Five articles, three frameworks, one team. Time to put it all in one place — the templates, the metrics, and the decision rules that actually stuck for the QuantumAPI team after 6 months. Copy, adjust, ship.

#devops #ai #sdd #atlas #gotcha #three-ways #playbook #dora

Apr 15, 2026 three-ways-ai

SDD, ATLAS, GOTCHA: When to Use What (And When They All Fail)

Four articles saying 'use this framework, it works'. Time for the uncomfortable part: none of them is a silver bullet. Each has a sweet spot and a cost — and teams that stack all three without thinking end up slower than they started.

#devops #ai #sdd #atlas #gotcha #three-ways #anti-patterns

Apr 14, 2026 three-ways-ai

Continuous Learning When AI Writes Half Your Code

SDD gave us small batches. GOTCHA gave us useful feedback. But when production broke at 2am, the on-call engineer could not explain the AI-generated code they were looking at. Time to use AI as a tutor, not just a generator — and finally close the Third Way.

#devops #ai #gotcha #learning #three-ways

Apr 13, 2026 three-ways-ai

AI Code Review That Doesn't Cry Wolf

SDD gave us small PRs. But the AI reviewer still left 22 comments on a clean 5-file diff, and 19 were noise. The team learned to ignore it. Here is how GOTCHA turns a crying-wolf reviewer into one that gets read.

#devops #ai #gotcha #code-review #mistral #three-ways

Apr 12, 2026 three-ways-ai

Specs Before Code: Why AI Needs SDD to Keep Batches Small

The 47-file PR from the last article was not the AI's fault. It was ours — we gave it a prompt, not a spec. In this article, the same team rescues the same feature using Spec-Driven Development, and watches lead time drop from 6 days to 1.

#devops #ai #sdd #spec-kit #atlas #three-ways

Apr 11, 2026 three-ways-ai

The DevOps Handbook Turns 10. AI Just Broke Half of It.

Ten years after Gene Kim's book, AI was supposed to finally unlock The Three Ways. Instead, most teams are shipping bigger batches, noisier feedback, and shallower learning. Here is why — and what to do about it.

#devops #ai #sdd #atlas #gotcha #three-ways

Apr 9, 2026 atlas-gotcha

A Scaffold Generator for Claude — Built on Gartner's Seven Invariants

Starting a new project with Claude is the moment everything goes wrong. No guardrails, no hooks, no context. So I built a scaffold generator that asks the right questions, sets up 14 security hooks before you write a single line, and follows Gartner's seven invariants for governing Claude Code at scale.

#claude-code #scaffolding #security #governance #gartner #atlas #gotcha #hooks

Apr 4, 2026 infra-hub

The Infrastructure Hub Reference Architecture

Eight articles, six new plugins, three new database tables. The complete map of the Infrastructure Hub — how it extends the AI-Native IDP, how QuantumAPI secures it end to end, and how to deploy it for sovereign or air-gapped environments.

#platform-engineering #backstage #architecture #terraform #kubernetes #quantumapi #reference

Apr 3, 2026 atlas-gotcha

ATLAS + GOTCHA vs Spec-Driven Development: What's the Same, What's Different

Spec-Driven Development is everywhere in 2026. GitHub spec-kit, Amazon Kiro, Thoughtworks radar. But ATLAS + GOTCHA has been doing structured AI development since before SDD had a name. Here's an honest comparison — what's shared, what's different, and when to use each.

#ai #atlas #gotcha #sdd #spec-driven #methodology #prompts

Apr 3, 2026 infra-hub

Drift Detection: When Your Infrastructure Doesn't Match Your Code

Someone clicked in the Azure portal. A manual change. A hotfix that nobody committed. Now your Terraform state says one thing and reality says another. Here's how to detect drift automatically, explain it with AI, and fix it from Backstage.

#platform-engineering #backstage #terraform #drift #infrastructure #ai

Apr 2, 2026 infra-hub

Chat with Your Infrastructure

You have 40 Terraform modules across 5 clients. You can't remember which AKS version client ACME runs. Here's a conversational panel in Backstage that reads your catalog, your state, and your docs before answering — so you stop asking Slack and start asking your platform.

#platform-engineering #backstage #ai #chat #infrastructure #rag

Apr 1, 2026 infra-hub

CAB Automation: Signed Approvals and Compliance Evidence

Article 4 built the CAB workflow. Now we make it audit-proof: every approval signed with ML-DSA, every evidence package cryptographically sealed, and compliance reports generated automatically for NIS2 and SOC2 auditors.

#platform-engineering #backstage #change-management #post-quantum #compliance #cab #quantumapi

Mar 31, 2026 infra-hub

Secrets and Post-Quantum Identities for Infrastructure

Your Terraform state has connection strings. Your pipelines have API keys. Your SSH keys are RSA. Here's how to manage infrastructure secrets with QuantumVault, rotate credentials automatically, and use AI to detect secret sprawl across your modules.

#platform-engineering #backstage #security #post-quantum #quantumvault #quantumid #terraform #secrets

Mar 30, 2026 infra-hub

Pipelines from Backstage — AI, Change Requests, and the Enterprise Reality

Your team runs pipelines in Azure DevOps, GitHub Actions, and GitLab CI. But the real problem isn't three UIs — it's that every production change needs a Change Request, a CAB review, and evidence that nobody has time to prepare. Here's how AI turns your pipelines into an enterprise-ready deployment machine.

#platform-engineering #backstage #pipelines #ci-cd #azure-devops #github-actions #gitlab-ci #change-management #ai

Mar 29, 2026 infra-hub

Multi-tenant Infrastructure — One Platform, Many Clients

Your company manages infrastructure for 15 clients. Each one has different modules, different clouds, different approval processes. You can't keep track of it all in spreadsheets. Here's how to make Backstage work for both internal DevOps teams and managed services providers.

#platform-engineering #backstage #multi-tenant #msp #infrastructure

Mar 28, 2026 infra-hub

Golden Path Terraform Modules

Every team creates Terraform modules differently. Different folder structures, different testing approaches, different ways to document inputs. Here's how to build standard module templates in Backstage for Azure, Scaleway, AWS, and GCP — so every new module starts right.

#platform-engineering #backstage #terraform #golden-path #scaffolder

Mar 27, 2026 infra-hub

Your Infrastructure Has No Catalog

Your Terraform modules are scattered across repos. Nobody knows which version to use, who owns what, or if that module from 2024 still works. Here's how to build an infrastructure catalog in Backstage — the same way you'd catalog services, but for modules, patterns, and environments.

#platform-engineering #backstage #terraform #infrastructure #catalog

Mar 26, 2026 atlas-gotcha

Your 90-Day Roadmap: From Developer to ATLAS Architect

The series finale. A self-assessment, a 90-day plan, and everything you need to make ATLAS + GOTCHA a permanent part of how you work.

#ai #atlas #gotcha #career #roadmap #architecture

Mar 26, 2026 ai-native-idp

TechDocs RAG — AI That Reads Your Docs

Your AI assistant can write code but can't answer 'what's our retry policy for Service Bus?' Here's how to index your platform documentation in a vector store and give AI real answers from real docs.

#platform-engineering #backstage #ai #rag #pgvector #techdocs

Mar 26, 2026 ai-native-idp

The Reference Architecture

Seven articles, six AI plugins, one .NET service, one Backstage portal. The complete map — how every piece connects, how to deploy it, and how to extend it with your own AI features.

#platform-engineering #backstage #ai #architecture #kubernetes #reference

Mar 25, 2026 ai-native-idp

AI-Assisted Incident Response

At 3am the invoice-api returns 500 errors. The on-call engineer opens a wall of logs. Here's a Backstage plugin that reads the catalog, checks recent deployments, and suggests what went wrong — before you finish reading the first stack trace.

#platform-engineering #backstage #ai #incident-response #observability

Mar 24, 2026 ai-native-idp

The AI Governance Dashboard

You've added AI to your platform. Catalog enrichment, scaffolding, code review, RAG. Now who controls it? Here's a Backstage plugin that gives the platform team visibility and control over every AI feature.

#platform-engineering #backstage #ai #governance #observability

Mar 23, 2026 ai-native-idp

The AI Code Review Plugin

Code reviews without context are guesswork. Here's a Backstage plugin that reads the service catalog before reviewing a PR — so it knows the architecture, the dependencies, and the rules before it reads the first line of code.

#platform-engineering #backstage #ai #code-review #github #mistral

Mar 20, 2026 ai-native-idp

Teaching Your Catalog to Think

Your service catalog is wrong. Not because someone made a mistake — because nobody has time to keep it right. Here's a Backstage plugin that reads your code and updates the catalog automatically.

#platform-engineering #backstage #ai #service-catalog #mistral

Mar 20, 2026 ai-native-idp

AI-Powered Software Templates

Static templates generate the same project every time. Here's a Backstage scaffolder where developers describe what they need in plain English and get a project with the right dependencies, configuration, and an ATLAS+GOTCHA prompt ready to go.

#platform-engineering #backstage #ai #scaffolder #atlas #gotcha

Mar 14, 2026 ai-native-idp

Why Your IDP Doesn't Help

Most Internal Developer Platforms become graveyards of outdated templates and docs nobody reads. Here's why — and what changes when AI becomes part of the platform itself.

#platform-engineering #backstage #ai #developer-experience #idp

Mar 13, 2026 atlas-gotcha

Advanced Patterns and the Top 10 Anti-Gotchas

The most common mistakes developers make when using AI for architecture — and how ATLAS + GOTCHA prevents them. Plus 20 ready-to-use prompt templates.

#ai #atlas #gotcha #patterns #prompts #architecture

Mar 11, 2026 atlas-gotcha

Hands-On: A Production CI/CD Pipeline with Security Scanning

Build an Azure DevOps pipeline for your .NET API — with SAST, container scanning, and zero-downtime deployment to AKS. Designed with ATLAS + GOTCHA.

#ai #atlas #gotcha #azure-devops #cicd #kubernetes #security

Mar 9, 2026 atlas-gotcha

The Perfect Mapping: ATLAS Meets GOTCHA

How every human decision in ATLAS maps to an AI instruction in GOTCHA — plus the master prompt template.

#ai #atlas #gotcha #mapping #prompts

Mar 9, 2026 atlas-gotcha

Hands-On: Build a Users API in 45 Minutes with ATLAS + GOTCHA

From checklist to production. A complete .NET users API with PostgreSQL, JWT, and Kubernetes — built using the ATLAS + GOTCHA workflow.

#ai #atlas #gotcha #dotnet #postgresql #kubernetes #jwt

Mar 9, 2026 quantum-safe-cloud

The Quantum-Safe Reference Architecture

Series finale. The complete architecture combining QuantumVault, QuantumAPI EaaS, and QuantumID — with a full ATLAS document, a GOTCHA prompt, and a reference diagram you can bring to your team.

#security #post-quantum #architecture #quantumapi #azure #kubernetes #reference

Mar 5, 2026 atlas-gotcha

GOTCHA: The 6 Layers That Make AI Think

A deep dive into the GOTCHA framework — six layers that turn your thinking into precise AI instructions.

#ai #gotcha #prompts #framework

Mar 5, 2026 quantum-safe-cloud

Zero Trust in AKS: mTLS with Quantum-Safe Certificates

Service-to-service communication in Kubernetes with no implicit trust. How to use QuantumAPI to issue ML-DSA certificates and enforce mutual TLS between your microservices.

#security #post-quantum #zero-trust #kubernetes #mtls #quantumapi #aks

Mar 1, 2026 atlas-gotcha

ATLAS: The 5 Phases Every Architect Must Know

A deep dive into the ATLAS checklist — five phases that structure your thinking before you ever talk to an AI.

#ai #atlas #architecture #checklist

Mar 1, 2026 quantum-safe-cloud

Quantum-Safe CI/CD: No More Secrets in Pipelines

Supply chain attacks, secrets in YAML, and unsigned images. How to extend your Azure DevOps pipeline with QuantumVault secrets and Cosign image signing.

#security #post-quantum #cicd #azure-devops #quantumvault #kubernetes #supply-chain

Feb 26, 2026 atlas-gotcha

Why AI Prompts Fail and What to Do About It

The chaos of unstructured AI prompts. Introducing ATLAS + GOTCHA as a structured approach to AI-assisted development.

#ai #atlas #gotcha #prompts

Feb 25, 2026 quantum-safe-cloud

QuantumID: Authentication Built for What's Coming

JWT weaknesses you should know about, and how to replace custom JWT auth with QuantumID — an OIDC provider that signs tokens with ML-DSA instead of RS256.

#security #post-quantum #authentication #oidc #quantumid #dotnet

Feb 22, 2026 quantum-safe-cloud

QuantumAPI EaaS: Encrypt Anything

How hybrid ML-KEM + AES-256-GCM encryption works, and how to use it to protect sensitive fields in your database without managing cryptographic keys yourself.

#security #post-quantum #encryption #quantumapi #dotnet #postgresql

Feb 18, 2026 quantum-safe-cloud

QuantumVault: Secrets That Can't Be Stolen

Where secrets actually live in most systems — and why it's a problem. How QuantumVault solves secret management with ML-KEM encryption and QRNG key generation.

#security #post-quantum #quantumvault #secrets #azure #dotnet

Feb 15, 2026 quantum-safe-cloud

Why Your Encryption Has an Expiry Date

RSA and elliptic curves will break when quantum computers arrive. Attackers are already collecting encrypted data today. Here's what that means for your systems, and what to do about it.

#security #post-quantum #cryptography #quantumapi

Jan 31, 2026 ai-in-production

Production Readiness: The Complete AI Feature Checklist

Seven articles of patterns, principles, and code. Now the question is simple: is your AI feature actually ready to ship? Here's the complete checklist — one per domain, covering everything from resilience to compliance.

#ai #architecture #production #checklist #dotnet

Jan 27, 2026 ai-in-production

Testing AI Features: Reliable Tests for an Unreliable Component

You can't assert response == 'expected string'. Here's how to test AI features in three layers — unit tests that don't call the model, integration tests for failure paths, and eval tests that measure quality without exact matching.

#ai #architecture #testing #dotnet #quality

Jan 24, 2026 ai-in-production

Integrating AI into Existing Systems: Patterns That Don't Break What Works

You have a production API with real users. Here's how to add AI features to it incrementally — without blocking requests, without breaking existing data, and without a big-bang rewrite.

#ai #architecture #integration #dotnet #patterns

Jan 22, 2026 ai-in-production

Governance and Compliance: What Legal Will Ask Before Your AI Goes Live

Personal data, GDPR, audit logs, consent, the EU AI Act. Here's how to design AI systems that survive a compliance review — before Legal comes knocking.

#ai #architecture #gdpr #compliance #security

Jan 19, 2026 ai-in-production

The Cost Problem: How to Keep Your AI Bill Predictable

Token costs are invisible until they aren't. Here's how to design for predictable AI spending — with caching, input limits, model routing, and budget guardrails.

#ai #architecture #cost #caching #dotnet

Jan 16, 2026 ai-in-production

Observability for AI Systems: Measuring What You Can't See

HTTP 200 doesn't mean the AI is working. Here's how to measure latency, quality, cost, and drift in real time — before users find the problems.

#ai #architecture #observability #opentelemetry #dotnet

Jan 13, 2026 ai-in-production

Designing for Failure: What Happens When the AI Doesn't Work

AI APIs go down, time out, and return garbage. Here's how to build systems that handle it — with fallbacks, circuit breakers, and degraded mode.

#ai #architecture #resilience #dotnet #polly

Jan 10, 2026 ai-in-production

The Gap Nobody Talks About: From AI Demo to AI in Production

Every AI demo works. Most AI systems fail in production. This is about the gap between the two — and how architects think about it differently.

#ai #architecture #production #reliability