SDLC blueprint

DevOps — connection to the SDLC foundation

DevOps treats the **entire** delivery pipeline — from code commit to production operation — as a **single system** to be optimized. The blueprint's **A–F** phases still apply; DevOps adds **automation

DevOps — connection to the SDLC foundation

DevOps treats the entire delivery pipeline — from code commit to production operation — as a single system to be optimized. The blueprint's A–F phases still apply; DevOps adds automation, monitoring, and operational feedback as first-class concerns.

1. SDLC phases A–F (DevOps lens)

Phase DevOps expression Notes
A — Shape Feature flagging strategy; infrastructure requirements alongside functional requirements Consider operability from the start
B — Plan Pipeline design; deployment strategy; SLO targets Plan how you will deploy and monitor, not just what you will build
C — Build CI: automated build, lint, unit tests on every commit; IaC for environments "Works on my machine" is not acceptable
D — Verify Automated testing pyramid; security scanning; performance testing in pipeline Shift quality left; automated gates replace manual checkpoints
E — Release CD: automated deployment; canary/blue-green releases; feature flags Decouple deployment from release; make deployments boring
F — Operate & learn Monitoring, alerting, incident response, post-mortems; production feedback → A Close the loop; production is not "someone else's problem"

Prescriptive rule: DevOps is not just "automate deployment." It requires cultural change: shared ownership between Dev and Ops, blameless learning, and continuous improvement of the entire pipeline.

DevOps teams maintain the blueprint tracking spine with pipeline emphasis:

Artifact DevOps mapping
Intent / request Feature request, bug report, or operational improvement
Spec Acceptance criteria + operability requirements (SLOs, monitoring, rollback)
Plan Pipeline configuration; deployment plan; runbook
Tasks Development + infrastructure tasks; pipeline improvements
PRs Code + IaC + pipeline config changes in the same review
Reviews Code review + deployment review + security review
Release Automated deployment; release notes; monitoring confirmation

Prescriptive rule: Infrastructure and pipeline changes should go through the same review process as application code. "ClickOps" (manual infrastructure changes) is technical debt.

3. Ceremony intents (C1–C6) ↔ DevOps practices

Intent DevOps practice Notes
C1 — Align & decide Architecture review with operability lens; SLO setting Include Ops perspective in design decisions
C2 — Plan the slice Sprint/iteration planning includes deployment and monitoring tasks Operational readiness is part of "done"
C3 — Execute & unblock Stand-up (includes pipeline health); on-call coordination Pipeline failures are blockers, not afterthoughts
C4 — Review & quality Deployment review; DORA metrics review; SLO review Measure the pipeline, not just the product
C5 — Reflect & improve Blameless post-mortem; pipeline retrospective Learn from incidents and near-misses
C6 — Knowledge share Runbook updates; incident learnings; monitoring dashboards Operational knowledge is team knowledge

4. Role archetypes (blueprint hats on a DevOps team)

DevOps role Typical archetype emphasis Notes
SRE (Site Reliability Engineer) Assure & ship + Build Reliability engineering; SLOs; error budgets; automation
Platform engineer Build + Assure Internal developer platform; CI/CD infrastructure; developer experience
Release engineer Assure & ship + Orchestrator Release process; deployment automation; rollback procedures
Development team Build (primary) + Assure (shared ownership) "You build it, you run it" — shared production accountability
On-call engineer Assure (incident response) Rotated among team; first responder for production issues

5. What DevOps adds beyond the foundation

  • Automation — CI/CD pipelines, IaC, automated testing as first-class engineering.
  • Shared ownership — "you build it, you run it" culture.
  • Feedback loops — production monitoring feeds development priorities.
  • DORA metrics — research-backed measures of delivery performance.
  • Blameless culture — incidents are learning opportunities, not blame opportunities.

6. Anti-patterns (prescriptive "don't")

Anti-pattern Fix
"DevOps team" as a separate silo DevOps is a culture, not a team name; embed practices across all teams
Automating a broken process Fix the process first, then automate; automation amplifies both efficiency and dysfunction
No production ownership by developers "You build it, you run it" — developers participate in on-call and incident response
Measuring only deployment frequency Use all four DORA metrics; frequency without quality is just faster failure

7. References in-repo

Canonical source

Edit https://github.com/autowww/blueprints/blob/main/sdlc/methodologies/devops/foundation-connection.md first; regenerate with docs/build-handbook.py.