The Future of a Programmer

I am a programmer. I love programming. I prefer facing a computer over facing humans (hahaha 😂), because computers give immediate feedback: TRUE (1) or FALSE (0), with nothing in between. I’m fortunate to have started my career in an era when programmers thrived.

AI was created by programmers (technologists are programmers too). We envisioned AI freeing us from repetitive, low-value tasks, like household chores (robot vacuums), transportation (self-driving cars), and more, so we could spend time on meaningful work, family, and improving human life.

But the rise of Generative AI (GenAI) has disrupted that vision. Instead of handling mundane tasks, AI is now encroaching on creative and analytical roles, including programming. Ironically, we may soon find ourselves doing the chores we hoped AI would take over, while AI writes the code.

What Lies Ahead for Programmers?

First, prompt engineering, or more accurately, prompt literacy, should become a mandatory skill, as essential as Word, Excel, and PowerPoint are today.

According to AI-2027.com, Superhuman Coders will emerge by 2027. Does this mean the end of programmers? Yes, unless we upskill and reskill. The industry won’t need as many traditional coders. Instead, coding responsibilities will shift:

Architects (Solution & Technical): Strategists and Orchestrators of AI-Driven Development

Architects will evolve into AI Orchestrators, leveraging AI’s remarkable strengths in research, analysis, comparison, and recommendation. While AI excels at generating insights and justifying its recommendations, it's the architect who defines the critical guardrails, blending deep technical acumen with irreplaceable human judgment. A key challenge for architects will be understanding the "black box" nature of some AI models, discerning when and how to fully trust AI-generated architectural designs, and rigorously validating them against real-world constraints and organizational policies.

• Experiential Wisdom: Translating vague business requirements into precise, actionable AI instructions (e.g., "design a serverless architecture for processing 10 million streaming logs per hour, balancing performance and cost").
• Tacit Knowledge: Filtering AI recommendations with nuanced insights, such as, "the proposed architecture must be scalable to handle 100 million transactions per hour during peak events, accounting for future growth."
• Ethical Foresight: Ensuring AI-generated recommendations align with stringent regulatory requirements, such as GDPR data transfer stipulations (e.g., "ensure the processing of EU data subjects occurs exclusively on servers within the EU, or with approved data transfer mechanisms if in Singapore, whilst balancing performance and csot").

Business Analysts (Now encompassing Coding Manager responsibilities): Bridging AI Capabilities and Human Needs

Business Analysts will become the essential link in the AI-to-human translation layer. Beyond their current scope of capturing and documenting user requirements, their role expands significantly. They will also take on "AI Product Ownership," responsible not only for the user experience of the application itself but also for how users intuitively interact with the AI features within it, ensuring effectiveness and adoption.

• Prompt Refinement: Expertly converting high-level stakeholder requests like "Improve UI consistency" into specific, actionable prompts for AI (e.g., "Re-arrange buttons for visual hierarchy and adherence to brand guidelines across all modules").
• Hyper-Personalization: Moving beyond static, configurable dashboards to drive truly personalised applications. For instance, designing an ERP system UI and workflows tailored specifically for a regional office with only three staff members, or a finance application that dynamically adjusts its user experience based on the user’s cognitive style and preferences.
• Practical Adaptation: While AI can generate initial code, humans, especially Business Analysts, will be crucial in refining and adapting it for real-world scenarios (e.g., adding a new, complex approval workflow that requires specific agency sign-offs).

Quality Assurance: Guardians of Application Integrity, Including AI-Generated Code

QA professionals are the last line of defense, and their guardianship of applications will significantly expand to encompass AI-generated code. This includes rigorous technical audits, ensuring compliance with both technical standards and evolving AI-specific regulations. Their expanded role will be to ensure AI’s efficiency never compromises safety, fairness, or reliability. This demands new methodologies for testing AI models themselves, beyond merely the code they produce. This will involve a deeper understanding of model explainability (XAI) and robustness against adversarial attacks, which are crucial for high-stakes AI applications.

• AI Governance: Ensuring AI ethics are inherently built into every application, upholding the fundamental AI principle of "Do No Harm." This also involves auditing AI outputs for potential bias, identifying security risks unique to AI, and ensuring unwavering compliance with regulatory frameworks (like GDPR, ISO standards, and industry-specific regulations).
• Technical Safeguards: Implementing rigorous code reviews, enhancing test automation to cover comprehensive edge cases (e.g., "Verify payment API fails gracefully during unexpected network drops and power outages").
• Lifecycle Oversight: Proactive monitoring of AI systems post-deployment to detect "model drift" or unexpected behavior (e.g., "Investigate why the loan approval model is suddenly rejecting a disproportionate number of rural applicants who previously qualified").

Closing

AI is here today.

In each Industrial Revolution, we have witnessed the diminishing of old roles and the creation of new ones. Each industrial revolution has propelled humanity forward, not just technologically, but in living standards, health, and prosperity. AI, as part of IR4.0, may begin with disruption but will ultimately elevate human potential.

AI won’t erase programmers; it will reshape our value. The future belongs to those who:

• Architect AI’s direction with human wisdom,
• Manage its outputs for real-world impact,
• Guard against its risks with relentless scrutiny,
• And master many other new roles like CAIO, Transcriptionist, etc.

The mantra shifts from "Write code" to "Guide code", and that’s a future worth coding for.

In the future, could personalizing applications become a new norm?