About this article
Thank you for visiting this site. This article is a guide listing all 5 articles in the “Solution Architecture” category of the Architecture Crash Course for the Generative-AI Era series.
Solution architecture is the domain of design focused on solving a specific project’s problem. From requirements definition to quantifying non-functional requirements, ROI calculation, and PoC design — the craft of “winning with numbers, not technology” at the project level. If enterprise architecture is the city plan, solution architecture is the blueprint for an individual building.
Article index
1. Solution Architecture Overview — Win with Numbers, Not Technology
A bird’s-eye view of solution architecture: three-option comparison, non-functional requirements, ROI, and PoC. As the title “win with numbers, not technology” suggests, this is the entry point for developing the skill of translating technology decisions into business cases.
2. From Requirements to Design — What You Hear at a Desk Is the Tip of the Iceberg
Covers the craft of translating business requirements into technical design. Three-tier breakdown, MoSCoW prioritization, field observation, acceptance criteria — the practical techniques for capturing requirements you can’t catch from desk-based interviews alone.
3. Non-Functional Requirements — “Never Goes Down” Has No Price Tag
Covers the craft of defining availability, performance, scalability, operations, security, and migration as numbers. Also covers the IPA Non-Functional Requirements Grade and the relationship with SLA/SLO — putting a price tag on the vague demand of “it should never go down.”
4. Estimates and ROI — An Estimate That’s Exactly Right Is Usually a Lie
Covers TCO, three-point estimation, buffers, ROI, Payback, NPV, and handling qualitative benefits. Including the AI-era shift in estimation assumptions, learn how to assemble the numbers that get a proposal approved.
5. PoC Design — A PoC That Ends with “It Kinda Worked” Is Always a Failure
Covers Go/No-Go criteria, duration setting, the difference from MVP, AI PoC peculiarities, and weekly PoC cycles. Under the strict standard that a PoC that doesn’t produce a decision is always a failure, learn how to design effective PoCs.
Summary
This article listed all 5 articles in the Solution Architecture category of the Architecture Crash Course for the Generative-AI Era series.
Solution architecture is the most directly applicable knowledge for day-to-day projects. Beyond technical skill, learning the flow of quantifying requirements, translating them into business cases, and validating with PoCs dramatically increases your credibility as an architect.
For the overall series structure and other categories, see the master series index.
Hope you’ll check out the next article as well.