Previously, the developer profession often required memorizing numerous details—new programming languages, frameworks, and libraries—and this learning cycle never ended. While mastering one technology, specialists immediately prepared for the next.
However, artificial intelligence is beginning to disrupt this established process. Nevertheless, according to Aditi Nanda, Director of Ecosystem and Industry at Intel, developers should not stop learning; instead, they need to learn to be more selective about what they study.
Speaking at the DevSparks Chennai conference, Nanda noted that AI takes on a significant portion of routine work in software development, but this does not diminish the importance of fundamental engineering knowledge. On the contrary, it increases their value, as developers are increasingly responsible for defining what the AI should create and whether the final result is usable.
She compared these changes to the advent of GPS. Drivers no longer need to memorize every road, but they still must know when to accelerate, brake, or turn around if there is an obstacle ahead.
"You don't need to remember syntax. You don't need to remember frameworks. But you must decide what the outcome will be," she stated.
For developers, this means they can spend less time mastering each new language or framework and more time understanding the systems that these tools help build.
Nanda pointed out a common issue with generative AI: it can very quickly translate an idea into a working prototype, but deploying this prototype into a production environment requires a completely different set of skills.
"Use AI to achieve the first prototype, the demonstration, but do not stop studying your fundamentals. Do not stop studying your fundamental engineering principles," she emphasized.
These fundamentals go beyond simply writing code. Nanda stressed the necessity of understanding the market and the user problem even before making a technical decision. She said: "It is necessary to understand what the pain point of the market, consumer, or user is. We can create better solutions, but if no one uses them, it is a waste."
System knowledge is another area that, in her opinion, developers cannot afford to lose. Since AI-based applications move between cloud services, personal computers, peripheral devices, and data centers, developers must understand exactly where the load should be executed. For a corporate application, this decision can involve issues of cost, security, and compliance.
"Confidential HR information or financial information, for example, may be unacceptable to send outside the organization's environment. Thus, architecture remains the developer's responsibility, even when AI writes parts of the implementation," she explained.
The difference between a developer who understands AI and a developer who is completely dependent on AI becomes particularly important as AI makes software development accessible to people without traditional skills.
Nanda gave the example of using Cursor for quick application development while waiting for a flight. Creating a functionally complete application now takes negligible time compared to the past. However, preparing such an application for millions of users presents a different challenge.
"A professional will be responsible enough to understand how the user will view it. What errors are here? How can I test this? What is the compliance? What is the cost? Where will I store the data? What is the security? They will ask all these questions," she noted.
For Nanda, the future of the developer rests on three core competencies: understanding the 'why,' knowing the 'how,' and the ability to bring the solution to reality.
The first is understanding the user and the problem. The second is systemic and engineering knowledge, including architecture and trade-offs between cost, efficiency, and time-to-market. The third is implementation: testing, reliability, and ensuring the product works when it reaches the users.
She added that although AI can enhance each of these abilities, it cannot remove the responsibility from the person creating the product.
"Use AI as a tool. Use AI as an augmenting mechanism. Do not let it think for you," Nanda concluded.
For developers working in an industry where a framework might become obsolete before people have time to master it, this can be a valuable lesson: the skills worth keeping become increasingly valuable, even when the tools change.