"Programmers" Are Moving from People to Tools, Just Like "Computers" Once Did
GeekPark author Wildcard uses the historical analogy of the word "computer" shifting from a profession to a machine, framed around the October 24 Programmer's Day, to explain how AI coding tools are dissolving the identity boundaries of "programmers": it is not that jobs are disappearing, but that the name is being absorbed by the tool, while humans retreat to roles of verification, command, and sign-off. This judgment is supported by Anthropic's self-reported data, DHH's public statements, and data showing a contraction in entry-level positions.
"Computer" used to be a profession. In the late 19th century, Harvard Observatory hired a group of women to analyze star chart photographs one by one; they were called "Harvard Computers." During World War II, the National Advisory Committee for Aeronautics (NACA) in Langley, Virginia, employed hundreds of women to manually calculate wind tunnel data and flight trajectories. When ENIAC was built at the University of Pennsylvania in 1945, the six women responsible for getting the machine running were selected from among ballistic calculators—they became history's first "programmers."
Dorothy Vaughan, head of the West Computing Group, saw IBM mainframes arrive at Langley in the late 1950s. She taught herself FORTRAN and then led her team to learn it as well. As machines took over calculation, they turned into people writing programs for machines. The West Computing Group was disbanded in 1958. But human computers did not disappear overnight: when astronaut John Glenn prepared for his orbital flight aboard "Friendship 7" in 1962, the trajectory had already been calculated by an IBM computer, yet Glenn specifically requested that Katherine Johnson, a mathematician from the West Computing Group, manually verify the numbers. Machines handled the calculation; humans signed off on the results—an image almost identical to engineers reviewing agent outputs one by one today.
Later, the word "computer" remained, but what it referred to changed completely, from a group of women sitting at desks with slide rules to a machine. The profession was not declared extinct; its name was simply swallowed by its own tool. Today, no one celebrates "Computer Day," not because calculation has become unimportant, but because it is so ubiquitous that it no longer requires a dedicated group to represent it.
In this article published on GeekPark on October 26, 2026, author Wildcard uses this historical thread to explain what is happening now: the term "programmer" is slowly shifting from people to tools, just like "computer" once did.
This judgment is supported not by abstract predictions, but by a set of concrete data. Mid-month, Anthropic released a report on changes in its own workforce: from early 2025 to the report's release, the proportion of junior employees in the technology department dropped from 21% to 14%; among newly hired tech staff, the share of entry-level positions fell from 27% to 9%, a cut of two-thirds. In March this year, work hours for Claude agents within the company exceeded those of human employees for the first time; by August, the work hours directed by employees using agents were six times their own manual work hours. Backend data shows that interactive agent sessions are idle for about 80% of the time—not because the models run slowly, but because they are waiting for humans to provide the next instruction.
Writing code is no longer the bottleneck; human attention is. This is the picture outlined by Anthropic's self-reported data. It is not an independent audit result, but it comes from a company that has pushed AI programming to the extreme.
Another piece of evidence comes from public statements. In September this year, DHH, creator of Ruby on Rails, announced at the Rails World conference in Austin that his company, 37signals, would no longer use hand-written code as the standard method for software development. He named this initiative "Pencils down." The man who once said he would rather retire than hand the keyboard over to AI asked the audience how many people were still heavily hand-writing code; only five hands went up.
Werner Vogels, CTO of Amazon, offered a different perspective. In an interview in Shenzhen, he stated that Amazon's hiring of junior engineers remains "as before," noting that newcomers must learn to collaborate with others, understand systems that have been running for twenty years, and judge which AI-generated code requires careful inspection. He also introduced the concept of "verification debt": AI generates code, but if no one confirms whether it is correct, that uncertainty sits there like technical debt. Yet even in his narrative, the value of new hires has shifted from "writing code" to verification, collaboration, and judgment.
Macro data also points to the bottom rung of the ladder being pulled away. Stanford research shows that the number of software developers aged 22 to 25 in the United States is nearly 20% lower than the peak at the end of 2022, while senior developers remain largely stable. Domestically, data from Maimai shows demand for backend development positions dropped by 41%, while AI positions rose 7.9 times; recruitment for field deployment engineers increased 15 times year-over-year. What was pulled away was not the entire ladder, but the very bottom rung.
Returning to the date itself, October 24. It has never been an official holiday; it only became the conventional "Programmer's Day" around 2015, a code understood only by insiders. Growing up with it was a whole set of self-deprecation—"code farmer," "code slave," "code beast." The premise of such self-mockery is a threshold: earning significantly higher wages than peers and doing work incomprehensible to outsiders gives one the confidence to call oneself a migrant laborer.
But the code has left the circle. In 2020, CSDN held the first October 24 Programmer's Day in Changsha; by 2024, Weibo topic views reached 4.07 billion. This year, the conference was renamed the "October 24 Programmer Day & AI Builder Conference," with the theme "Everyone Can Build." Sitting at the opening ceremony were founders of several robotics companies. Many creating "new programmer works" have never hand-written a single line of code.
Co-optation expanded the festival's circle; AI made the circle so large it barely has boundaries. A festival open to everyone essentially belongs to no one.
Wildcard's judgment is: festivals usually do not die from being canceled; they die when no one needs them anymore to confirm who they are. Programmers are changing from "people who write code" to "people who review code," "people who assign tasks to AI," and "people who sign off on results." These jobs are all important, some arguably harder than writing code, but they share a common trait—they do not really need the insider code of "1024."
This analogy has a clear boundary: Dorothy Vaughan's generation did not hold onto the name "computer," but became the earliest programmers—the result of identity transfer was the birth of new professions, not the complete elimination of old ones. What future generations will call the person staring at six agent terminals simultaneously is currently unknown. What can be confirmed is that 1024 remains 2 to the power of 10; the number will not change, only who needs it to prove themselves will.