EVERYTHING
Business Insider reports that software engineering, once one of the most lucrative and in-demand roles in technology, is facing an irrevocable change due to AI advancements. Within just a few weeks late last year, new AI models from tech giants dramatically improved coding tools, making sophisticated tasks accessible almost overnight.
The Speed of Disruption
The impact of this shift has been immediate and profound for practicing engineers. Amy Surrett, an engineer in Greenville, South Carolina, experienced this firsthand when she used Anthropic's Claude Code to build a complex payment feature. She noted that while the project would have taken two or three days to code manually, Claude completed it in just over an hour. As Surrett stated, "It felt like the point of no return. This industry is not going to be the same. My job is not going to be the same."
This acceleration was noted by Andrej Karpathy, a former founding research scientist at OpenAI who later joined Anthropic. In February, he observed that it was difficult to communicate how much programming had changed due to AI in only two months, noting that before December, coding agents "basically didn't work."
The Scale of the Change
This technological surge is not limited to small projects; it affects entire corporate infrastructures. At tech giants like Google, AI is now reportedly writing as much as 75% of the company’s code. This transition has spawned a new lexicon of terms and driven hundreds of billions of dollars in AI investment globally.
The disruption highlights how coding, with its clearly defined rules, is highly susceptible to automation compared to other professions. The industry is now grappling with several core questions:
- What does it mean to be a coder if an application can be built without writing code?
- If engineers do not need to write code directly, what new opportunities will emerge?
- Will the developers who create and improve these AI tools automate themselves out of jobs altogether?
Existential Crises in Development
The pace of change has been both invigorating and destabilizing for many technical workers. Kent Dodds, a software engineer who previously taught coding skills, experienced his own "existential crisis" when he used Cursor, an AI-assisted coding tool. He set up an agent to handle the encryption and offline viewing features for his students' videos, which the AI "nailed it on first try." This single morning effectively evaporated weeks of potential work.
The industry is now entering a period of intense evaluation, moving beyond simple productivity gains. The lessons learned by engineers as they navigate this shift—from emotional impact to unlocking AI's full potential—are expected to be broadly relevant as the technology begins to disrupt other white-collar sectors.
Ultimately, the rise of sophisticated coding agents signals a massive structural overhaul in how software is created, moving the focus from manual execution to strategic oversight and prompt engineering.