Why Architectural Thinking Matters More Than Ever

I’ve recently been involved in devising a new process where any engineer or architect can share architectural proposals and bring their ideas to a group for discussion.

At first, this might seem like just another governance step. But it actually serves a much bigger purpose.

Another goal is to encourage everyone in engineering to think more architecturally.

With AI speeding up software development, being able to write code is no longer what sets people apart. Now, the real value comes from deciding what to build, weighing different options, understanding trade-offs, and making good technical choices.

These are architectural skills, and they matter for everyone - not just architects.

The Changing Nature of Engineering

For a long time, software engineering expertise was measured by how well someone could design and build complex solutions.

Now, AI tools can generate code, write tests, suggest ways to implement things, and speed up development in ways that seemed impossible just a few years back.

This opens up a great opportunity. Engineers can spend less time on repetitive tasks and more time on work that adds real value.

But AI shifts where the bottleneck is.

The main challenge isn’t just writing code anymore. Now, the challenge is picking the right solution.

AI can create an implementation, but it can’t take responsibility for decisions. It also can’t fully grasp the company’s context, business goals, rules, operational limits, or long-term plans.

That responsibility still belongs to engineers.

In many ways, the skills that used to set senior engineers apart are now things every engineer needs to learn:

• Understanding business value.
• Evaluating trade-offs.
• Reflecting on alternative approaches.
• Assessing risk.
• Thinking about maintainability and operational impact.
• Understanding long-term consequences of technical decisions.

Engineers will still need deep technical skills, but strong judgment will be just as important.

Encouraging Architectural Thinking

One of the best ways to build good judgment is by making decisions and talking them through with others.

A process for submitting architectural ideas can help encourage this.

This kind of process should give anyone the chance to share an architectural idea, question an existing approach, suggest a new standard, or get advice before making a big technical decision.

Typical examples include:

• Introducing a new technology, framework, or platform.
• Establishing a new engineering standard.
• Changing an architectural pattern.
• Solving a frequent technical problem.
• Improving reliability, security, maintainability, or developer experience.
• Exploring new AI or automation capabilities.

Instead of just presenting a solution, people submitting proposals are encouraged to think through some key questions:

• What are you proposing?
• Why is the change needed?
• What benefits would it deliver?
• What are the trade-offs?
• What decision or guidance are you seeking?

These questions help engineers look beyond just the details and think about the bigger picture.

Trade-offs are treated as a top priority.

Good architecture is rarely about perfect solutions. It’s about understanding compromises and making informed choices.

Learning Through Discussion

A technical review board can look at submitted proposals and consider different options, risks, and alternatives. Other experts can join the discussion when needed.

Whether the result is approval, feedback, more research, or spotting risks, the real value often comes from the discussion itself.

When engineers see how experienced architects and senior engineers review a proposal, they learn how architectural decisions are made.

Questions such as:

• What business problem are we solving?
• Is there a simpler approach?
• What operational burden does this introduce?
• What are the security implications?
• How does this affect long-term maintainability?
• Are there alternatives worth considering?

This helps build critical thinking skills throughout the organisation.

Creating a Record of Decisions

Another important part of the process is using Architecture Decision Records (ADRs).

ADRs capture:

• The problem being addressed.
• The options considered.
• The decision taken.
• The reasoning behind the decision.
• Any implementation guidance.

This way, knowledge isn’t lost when teams change or projects move forward.

Future engineers can see not just what decision was made, but also why it was made.

This is especially helpful in a world where AI assists us. As it gets easier to generate solutions, understanding the context and reasons behind past decisions becomes even more important.

What Else Can We Do?

A proposal process by itself won’t create great architects or improve engineering judgment.

To help everyone build these skills, we should keep encouraging practices like:

Design Reviews

Give engineers chances to present their solutions and get helpful feedback from their peers.

Exploring Alternatives

Encourage teams to write down several options before picking an approach, instead of just going with the first one that works. Documenting reasoning and trade-offs also helps.

Post-Incident Learning

Operational incidents often give us great chances to look at our decisions, assumptions, and any unexpected results.

Technical Communities

Architecture discussions, engineering forums, guilds, and sharing sessions help engineers see a wider range of perspectives.

Mentoring

Experienced engineers and architects can help newer colleagues learn how decisions are made and what factors matter.

Conclusion

AI will keep making software delivery faster and easier. That trend will probably speed up even more.

But while writing code gets easier, making good decisions doesn’t.

Being able to weigh trade-offs, test assumptions, understand consequences, and connect technical choices to business results will become even more valuable.

That’s why architectural thinking shouldn’t be just for architects. It’s becoming a core skill for all engineers.

Having an architectural proposal process helps build this skill across organisations. It lets every engineer share ideas, join in technical decisions, and grow the judgment that will shape successful engineering careers in the future.