About this article
Thank you for visiting this site. This article is a guide listing all 15 articles in the “DevOps Architecture (DevOps / SRE)” category of the Architecture Crash Course for the Generative-AI Era series.
DevOps architecture designs the entire flow of build, ship, and run. Under the philosophy that speed and stability can coexist, it systematically covers version control, CI/CD, testing, monitoring, and incident response. This is the largest category in the series, covering the full dev-ops lifecycle in 15 articles.
Article index
1. Overview — One Pipeline for Build, Ship, and Run
A bird’s-eye view of DevOps/SRE as a unified lifecycle. Covers the DORA four metrics, SLO + error budget, and AI-era machine-readable operational data — the map of the dev-ops lifecycle.
2. DevOps & SRE Big Picture — Speed and Stability Can Coexist
Covers DevOps origins, the DevOps/SRE relationship, DORA four metrics, CALMS, SLO and error budgets, and Platform Engineering. Learn the philosophical core of reconciling speed and stability — with historical incidents woven in.
3. Version Control — Git + Monorepo + GitHub Flow Is the Default
Covers five axes: repo structure, branching strategy, merge method, tagging, and secrets management. Learn the 2026 default configuration management stack: Git / monorepo / GitHub Flow / Squash merge.
4. Dev Environment and Local Execution — First Commit in Half a Day
Covers docker-compose, devcontainer, Codespaces, secrets distribution, and handling production data. Learn the dev-environment design that makes “a new hire’s first commit in half a day” a reality.
5. Code Review — PR 300 Lines + 1 Approver + CODEOWNERS
Covers PR granularity, review perspectives, approval rules, Conventional Commits, Squash merge, and AI-review preprocessing. Learn a code review operation that focuses human time on design decisions.
6. Test Design — Pyramid + Testcontainers + Branch Coverage
Covers the test pyramid, coverage targets, TDD, flake countermeasures, Testcontainers, and mock boundaries. Learn how to build a system that notices when things break — including AI-era test-first workflows.
7. CI/CD — GitHub Actions + OIDC + Feature Flag Is the Default
Covers staged CI/CD in practice, sub-10-minute PR builds, DB migration expand/contract, and IaC/GitOps/DevSecOps integration. Learn design that turns deploy speed into competitive advantage.
8. Deploy Strategy — Increase Frequency to Decrease Risk
Compares Rolling, Blue-Green, Canary, Feature Flag, Progressive Delivery, and automatic rollback. Learn the safe deploy design that prevents Knight Capital-class incidents.
9. Monitoring and Observability — Three Pillars + OpenTelemetry + SLO Alerts
Covers the three pillars (Metrics/Logs/Traces), the OpenTelemetry standard, Prometheus/Grafana/Datadog selection, and SLO burn-rate alerts. Also covers structured data design for AI-assisted diagnostics.
10. Log Design — Structured JSON + No PII + Phased Cold Storage
Covers log levels, structured JSON logging, required fields, collection and storage targets, retention periods, sampling, and PII protection. Also covers the new perspective of designing logs as messages for AI.
11. SLO and SLI — Don’t Aim for 100%; Buy Speed with Error Budget
Covers SLI/SLO/SLA differences, typical SLIs, availability benchmarks, error budgets, and burn-rate alerts. Learn the operational philosophy of using error budget to buy development speed under the mindset of “don’t aim for 100%.“
12. Incident Response — Resolve with Systems, Not Heroes
Covers on-call structure, SEV levels, command center, Blameless postmortems, Runbooks, and AIOps. Learn an incident operation that resolves issues through systems, not individual heroes.
13. SRE Practices — Toil Reduction and Chaos Engineering
Covers the key SRE practices: Toil reduction, error-budget operations, Chaos Engineering, PRR, and Platform Engineering. Practical enough for the field, including AI-era AIOps collaboration.
14. Documentation — README + ADR + OpenAPI in Git
Covers README, ADR, OpenAPI, Mermaid, docs-as-code, and postmortems. Learn how to manage documentation that AI can read and that doesn’t rot after six months by keeping it in Git.
15. Ticket & Project Management — Epic / Story / Task + 1-Day Granularity
Covers ticket tool selection, the Epic/Story/Task three-tier hierarchy, 1-day granularity, sprints, priority, and velocity. Learn an operation that makes tickets function as the organization’s memory.
Summary
This article listed all 15 articles in the DevOps Architecture category of the Architecture Crash Course for the Generative-AI Era series.
DevOps architecture is the largest category in the series. Not “build it and you’re done” but “keep it running” — including operations in the scope of design is indispensable in modern software development. Get the big picture from the overview, then start with the theme closest to your team’s current challenges.
For the overall series structure and other categories, see the master series index.
Hope you’ll check out the next article as well.