Why architecting is becoming more important than coding in the era of AI

For students of Engineering, memorise long lines of complicated syntax is no longer important. What is required is system design thinking

For the longest time, aspiring engineers believed that if they could learn and master the art of coding, they would have multiple career opportunities. While this is largely true, the advent of AI has changed certain things. Syntax that once took an entire semester to master can now be generated in seconds by an AI model. What cannot be generated is judgment: the ability to decide what to build, why it matters, and how the pieces of a complex system fit together. This is the shift every Engineering student needs to internalise: Coding, as a standalone skill, is being commoditised. Architecting is becoming the premium.

Systems thinking

Today, AI tools write boilerplate, autocomplete functions, translate pseudocode into working programmes, and even debug in real time. For students, this implies that the traditional method and headache of memorising long and complicated syntax and solving coding puzzles are no longer the real challenge. The premium skill is systems thinking, or the ability to zoom out from a single function to the architecture of an entire product.

Instead of being threatened by AI, students need to start viewing it as a partner who will guide them to prioritise the output and larger vision. If used correctly, one engineer can work with AI to prototype, test, and iterate at a pace that previously required a small set of people to work together as a team. But this will only work when the engineer giving prompts to AI has in-depth knowledge and grasp of architecture, scalability, design principles, and security to comprehensively understand the output generated.

This adoption needs to be accepted. There is an increased need for engineers to be fluent in prompting, analysing, reviewing, and refining AI-based outputs, just as the earlier generation was fluent in writing long, clean code by hand. Verification, critical evaluation, and the ability to spot when an AI-generated answer is subtly wrong are now core Engineering skills.

At the workplace

Recruiters are no longer interested in candidates who can simply code and memorise long lines of complicated syntax. Increasingly, they want people who have a sense of the product and an understanding of system design thinking, how well a candidate can work fluently with AI, and successfully incorporate their own voice in their projects. Technical interviews are based more on understanding a candidate's architectural knowledge and less on closed-form algorithmic puzzles.

The key takeaway for students is that they must prioritise building a portfolio that effectively showcases their architectural vision and reasoning, rather than just coding abilities. They must take part in and contribute to projects where they own and execute their own ideas and invest in the conceptual foundations -- data structures, distributed systems, security principles -- that let them judge whether AI's suggestion is actually sound.

Coding, in the narrow sense of writing syntax, is fading as a differentiator. But Engineering itself is not dying; it is maturing. Students who recognise this early and train themselves to think as architects rather than typists will thrive in tomorrow's workplaces.

The writer is Pro Vice Chancellor -Technology and Science, Manipal Academy of Higher Education.