AI is removing the entry-level work that trains tomorrow's senior engineers
Researchers at Seoul National University interviewed 14 software engineers in South Korea about who should get routine work now that AI tools can do it. Half were seniors deciding where tasks go, and the rest were students about to enter a job market that has stopped hiring them. Korea is a useful test case: work-related AI usage there reaches 51.8 percent, close to double the U.S. rate, and major technology firms have shut down open recruitment of junior developers.
One founder with more than twelve years as a developer said he let many junior engineers go last year and saw no impact whatsoever. Yet he also credited his own career to early grunt work: writing code as told, debugging daily, and checking and testing results. The article argues that minor bug fixes and routine debugging were never given to juniors because the output was valuable. They were given because doing them badly, then having a senior explain what went wrong, is how judgment develops. AI tools have absorbed that mechanism.
There is also a hidden cost. A senior at a mid-sized game company noted that generating code with AI is very cheap, but verifying it is very expensive. If fewer juniors become seniors, the people who can verify AI-written code may become scarcer over time.