A software engineering degree from Stanford used to be a golden ticket. Artificial intelligence has demoted it to bronze status, recent graduates say.
Elite students are shocked by the lack of job offers as they complete their degrees at what is often ranked as America’s best university.
When they were freshmen, ChatGPT had not yet been released to the world. Today, AI can code better than most humans.
Cutting-edge tech companies simply don’t need that many new graduates.
“Stanford computer science graduates have a hard time finding entry-level jobs” with top tech brands, said Jan Liphardt, an associate professor of bioengineering at Stanford University. “I think it’s crazy.”
While the rapid evolution of generative AI coding capabilities has made experienced engineers more productive, it has also hampered the job prospects of early-career software engineers.
The Stanford students describe a suddenly skewed job market, where only a small portion of graduates — those considered “cracked engineers” who already have thick resumes in product building and research — get the few good jobs, leaving everyone else to fight for the scraps.
“There’s definitely a very gloomy vibe on campus,” said a recent computer science graduate who asked to remain anonymous so he could speak freely. “People (who are) looking for a job are very stressed and it is very difficult for them to find a job.”
The upheaval is being felt at California universities, including UC Berkeley, USC and others. The job search is even more difficult for those with less prestigious degrees.
Eylul Akgul graduated from Loyola Marymount University last year with a degree in computer science. She didn’t get any offers, so she returned to Türkiye and gained some experience at a startup. In May, she returned to the United States, and yet she was “ghosted” by hundreds of employers.
“The programmer industry is becoming very oversaturated,” Akgul said.
The engineers’ biggest competitor is getting stronger day by day. When ChatGPT launched in 2022, it could only code for 30 seconds at a time. Today’s AI agents can code for hours, and complete basic programming faster with fewer errors.
The data suggests that while AI startups like OpenAI and Anthropic are hiring lots of people, that doesn’t make up for declining hiring elsewhere. Employment for specific groups, such as early-career software developers aged 22 to 25, has declined nearly 20% from its peak in late 2022, according to one study. Stanford study.
It is not only software engineers, but also customer service and accounting positions that are heavily exposed to competition from AI. The Stanford study estimated that entry-level hiring for jobs exposed to AI decreased by 13% compared to less exposed jobs such as nursing.
In the Los Angeles area, another study estimates that nearly 200,000 jobs are at risk. About 40% of tasks performed by call center workers, editors and personal finance experts could be automated and carried out by AI, according to a study. AI Exposure Index curated by resume builder MyPerfectResume.
Many startups and tech titans have not hesitated to announce that they are scaling back their hiring plans, as AI allows them to do more programming with fewer people.
Anthropic Chief Executive Dario Amodei said that 70 to 90 percent of the code for some of his company’s products is written by his company’s AI, called Claude. In May, he predicted that AI capabilities would increase to 50% of all entry-level white-collar jobs could be eliminated within five years.
A common sentiment among hiring managers is that where they previously needed ten engineers, they now only need “two qualified engineers and one of those with an LLM,” which can be just as productive, said Nenad Medvidović, a computer science professor at the University of Southern California.
“We don’t need junior developers anymore,” said Amr Awadallah, CEO of Vectara, an AI startup based in Palo Alto. “AI can now code better than the average junior developer from top schools.”
Certainly, AI is still far from causing the extinction of software engineers. While AI handles structured, repetitive tasks, human engineers jobs change towards surveillance.
Today’s AIs are powerful but “shredded”, meaning they can excel at certain math problems while failing basic logic tests and being inconsistent. A study found that AI tools slowed down the work of experienced developers by 19%, as they spent more time reviewing code and fixing errors.
Students should focus on learning how to manage and control AI work as well as gaining experience working with it, said John David N. Dionisio, a computer science professor at LMU.
Stanford students say they enter the job market and find a path to navigate; Competent AI engineers can find work, but basic, old-fashioned computer science jobs are disappearing.
As they encounter this surprise speed bump, some students lower their standards and join companies they wouldn’t have considered before. Some create their own startup. A large group of frustrated graduates decide to continue their education to beef up their resumes and learn more of the skills needed to compete with AI.
“If you look at the enrollment numbers over the last couple of years, they’ve skyrocketed for people wanting to do a master’s degree in the fifth year,” the Stanford graduate said. “It’s a whole different year, a whole different cycle for recruiting. I’d say half of my friends are still on campus in their fifth year of master’s degree.”
After four months of searching, Akgul, an LMU graduate, finally landed a job as a technical manager at a software consulting firm in Los Angeles. In her new job, she uses AI coding tools, but feels like she has to do the work of three developers.
Universities and students will need to rethink their programs and majors to ensure their four years of study prepare them for an AI world.
“This represents a dramatic reversal from three years ago, when all of my undergraduate mentees found great jobs at the companies around us,” Stanford’s Liphardt said. “That has changed.”