HarvTech

Docs / Decisions

Living document Updated 03 Jun 2026

Positioning — well-architected Azure landing zones

Walking back the generic-cloud framing in favour of a specific Azure landing-zone narrative, with AI as one of several modern workloads.

Context

The earlier hero positioning — “Secure Cloud platforms for AI and data workloads” — was deliberately generic about the cloud and led on AI. Both choices made sense at the time and are documented in the previous ADR, which this one supersedes.

Two things shifted the thinking:

  • Generic “Cloud” reads as soup-of-the-day SaaS. It avoids filtering out AWS/GCP conversations, but it also doesn’t earn a second look. “Secure Azure landing zone” is a concrete CAF concept that signals real depth of platform knowledge — a more honest signal of what I actually do day-to-day.
  • AI in the headline narrows the room. AI is a workload, not a discipline. Leading with it pushes away clients who want a well-architected platform for their workload — data, web, line-of-business, identity-heavy, whatever. The platform skills don’t change.

Same code in this repo. Same skills. Different framing.

Decision

  • Hero headline: “Well-architected Azure landing zones.”
  • Hero body: opens with “Secure foundations for modern workloads — including AI, data, and everything in between” before listing the day-job stack.
  • Meta description, footer, page titles: all rewritten around “well-architected Azure landing zones — secure foundations for modern workloads”.
  • Showcase in-progress card renamed from “Secure AI Landing Zone” to “Reference Azure landing zone”. AI (Azure OpenAI / Foundry) is described as one demonstrated workload on top, alongside data and app-tier examples — not the headline.
  • Writing page description and intro broaden from “AI workload security” to “modern-workload security (AI included)”.

What stays as it was

  • Capability cards (Cloud platform engineering, Terraform at scale, DevSecOps in CI/CD, etc.) were already generic — no changes needed.
  • The CV stays Azure-led with AWS/GCP as background; that was already accurate.
  • The data platform HLD still describes AI as a workload it could feed, where that’s factually true. The reframe is about headline positioning, not about scrubbing every accurate mention.

Consequences

  • Pro: specific framing earns trust. “Well-architected Azure landing zone” is a concept a CTO or platform lead recognises in three seconds.
  • Pro: broader workload appeal. Anyone shipping anything serious on Azure benefits from the same foundations; the AI angle becomes a demonstrated example rather than a gate.
  • Con: loses the deliberate “Cloud not Azure” hedging from the earlier ADR. The cross-cloud background is still in the body and CV, so the cost is small — a non-Azure prospect has to read one more sentence rather than bouncing on the headline.
  • Con: one ADR walking back another adds wiki noise. Kept the old ADR live and marked it superseded so the trade-off thinking is visible, rather than rewriting history.

Alternatives considered

  • Keep the generic “Cloud” framing. Rejected — too vague to differentiate, and the AI-headline side of the old framing was always the weaker half.
  • Lead with the AI specialism instead. Rejected for the workload-narrowing reason above. AI work remains in the showcase and writing, just not the headline.
  • Drop AI from the site entirely. Rejected — securing AI workloads (private endpoints around Foundry, identity around the model surface, WAF in front of the app tier) is genuinely interesting work and a credible portfolio piece, just not the lead claim.

Implementation

Single PR refactoring the hero, footer, meta description, showcase in-progress card, and writing page copy. Old ADR marked status: superseded with a forward link to this one. No code or Terraform changes — this is positioning, not architecture.