AGI Imminent: How To Survive An AI Software Engineering Job Apocalypse

Since the introduction of ChatGPT in November 2022, the industry has been speeding towards AGI (Artificial General Intelligence) – AI systems with human level intelligence and reasoning capabilities. More recently we’ve seen ChatGPT unveil their o1 model and Google has unveiled Gemini 2.5 pro, both of which now have longer-context comprehension, advanced reasoning and improved programming capabilities. Now, these Large Language Models (LLMs) can code, in some instances, more effectively than many human programmers when given relevant prompts as demonstrated here.

ASI On The Horizon

There is speculation around these LLMs becoming increasingly capable, so much so that they’ll be generating code and building systems from simplest of prompts and as this happens there will be less need for experienced programmers within companies potentially leaving many jobless. Prominent voices such as Eric Schmidt, Mark Zuckerberg and Bill Gates seem to support this notion with Schmidt stating in an address at the FFI Institute (Future Investment Initiative Institute) in October 2024 that this will likely happen within the next five years. He speaks of ASI (Artificial Super Intelligence) – an AI system that has the collective intelligence of every single human being on Earth combined.

But how likely is this? Many feel it is just hype.  Rhetoric designed to keep the AI hype train moving and investors investing – a quick search online reveals this sentiment! Perhaps it is! The truth is, we don’t believe that AI is going to replace our jobs. Quite the opposite, in fact. We believe that it’s actually going to make us developers even more productive and relevant.

Why AI Can’t Replace Programmers

There are things that even the best AI systems can’t do including deeply understanding an individual’s personal context and nuanced problem-solving, the ability to listen and read between the lines, and the capacity to effectively communicate complex ideas or solutions and gather intuitive feedback. Then there’s creativity – going beyond what was asked and delivering something more. These are just a few but you can already see why the ‘developer role’ is so important. If anything, AI can and should enhance the developer – making he/she more productive. it should assist in the development of solutions and products ensuring that fewer mistakes are made and that more features are comprehensively covered. This is exactly the primary function of tools like Github Copilot and Cursor. To that effect we’re still just at the beginning of a revolution in how developers code with AI. It’s exciting to see just what coding might look like in ten years’ time.

The Symbiotic Developer

For developers willing to adapt, this new era presents an unprecedented opportunity to become more relevant, productive, and ultimately, better at their craft.

Instead of viewing AI as a threat, you should embrace it as a tool to transform you into a better developer. You should be using AI assistants to generate code snippets, suggest improvements, fix bugs, explain code and much more. This automation of routine and time-consuming tasks will free you up to focus on higher-level problem-solving, architectural design, and creative solutions – the very aspects of software engineering where human ingenuity remains indispensable. The key is not to resist this change, but to learn how to effectively wield these AI-powered assistants to amplify your own skills.

So, how can developers not only survive but thrive in this AI-driven landscape and evolve into AI-specialist developers? Here are five solid points to guide your journey:

1. Master Prompt Engineering and AI Collaboration: Think of today’s LLMs as incredibly capable but literal-minded junior developers. Your ability to craft clear, concise, and context-rich prompts will directly impact the quality and relevance of the AI’s output. Learn how to effectively communicate your intent, break down complex problems into smaller, AI-manageable tasks, and iterate on AI-generated code. This includes understanding the limitations of current models, recognising potential biases, and knowing when to override or refine AI suggestions. Effective collaboration means guiding the AI, not blindly accepting its output. Want to learn Prompt Engineering? Take Google’s free Prompt Essentials course.
2. Develop Strong Code Review and Critical Thinking Skills: As AI takes on more of the initial code generation, the importance of rigorous code review and critical thinking will skyrocket. Developers will need to become adept at evaluating AI-produced code for correctness, efficiency, security vulnerabilities, and adherence to best practices and project-specific requirements. Understanding the “why” behind the code, not just the “what,” will be crucial. Your role shifts from pure code creation to that of a meticulous quality gatekeeper and architect.
3. Focus on System-Level Understanding and Architectural Design: While AI can generate functions or components, the ability to design robust, scalable, and maintainable software architectures remains a uniquely human skill. Developers should deepen their understanding of system design principles, software architecture patterns, and the integration of various components. AI can assist in exploring design options, but the strategic decisions and overarching vision will fall to the developer.
4. Become an Expert in AI-Powered Development Tools and Workflows: The software development toolkit is rapidly expanding with AI-integrated features. Familiarise yourself with AI coding assistants (like GitHub Copilot, Tabnine, or Amazon CodeWhisperer), AI-driven testing tools, and platforms that leverage AI for debugging and performance optimisation. Understand how to integrate these tools seamlessly into your development lifecycle (SDLC) to maximise efficiency, from initial design and coding to testing, deployment, and maintenance. This includes staying updated as these tools evolve at a rapid pace.
5. Cultivate Continuous Learning and Adaptability in AI Ethics and Best Practices: The field of AI is evolving at breakneck speed. Commit to lifelong learning, not just in new programming languages or frameworks, but specifically in the advancements and applications of AI in software engineering. Crucially, you should try to develop an understanding of the ethical implications of AI, including data privacy, bias in algorithms, and responsible AI development. An “AI-Advocate Developer” will be one who can leverage AI powerfully while also ensuring its application is ethical and sound.