- Fueled by the recent AI boom, companies are storming the campuses of top universities for scarce technical talent.
- These companies attract undergraduates and PhDs with exorbitant salaries and endless research funding.
- Now, students face a choice to leave noble academia and head for the promised land of AI.
On Friday afternoon, Cristóbal Eyzaguirre pulled out a Hot Wheels toy car and introduced Insider on Zoom. For the most part, it looks like a child’s toy, except it has “Tesla Model X” adorned all over the package.
The Stanford PhD student admitted that a toy car was his favorite toy out of all the goodies he got from private parties, job fairs, and Tesla events he attended as a plus-one. These events aim to attract young minds researching artificial intelligence to top companies.
“I really like Tesla’s,” said 27-year-old Eyzaguirre. “They really put in a lot of effort.”
Every decade brings a tipping point in technology that lures college students into the field. During the dotcom era, more students entered computer science. Entrepreneurship degrees rose as stories spread of highly successful startup founders in the on-demand economy. Then, during his Web3 craze over the past few years, students flocked to cryptocurrency and blockchain classes.
Students and recent graduates are now embracing AI as a home, and many hope to become early employees of the next great tech disruptor.
In interviews with insiders, a dozen university professors, students, recent graduates, and industry experts said that as more companies vie for AI supremacy, they raid college campuses looking for talent. says there is. And they hand out more than sweets and monogrammed water bottles to win the hearts and minds of their students. Their jobs offer mid-six figure salaries, the opportunity to work with top industry talent, and ample resources to tackle problems too expensive for academia to solve.
Recruitment has become so aggressive that some universities are seeing a drop in enrollment for AI PhDs. “They have a lot of resources and money hanging around [in front of] Kavita Barra, Dean of Computing and Information Sciences at Cornell University, said: “The student population is being wiped out.”
Students like Eyzaguirre should consider whether they are willing to turn their backs on virtuous academia in search of a place in the promised land of Silicon Valley.
Industry temptation
Hugging Face, a startup that offers ready-to-use machine learning models, has gotten a head start on the adoption cycle.
In early 2022, the company launched a series of university lectures, providing virtual demos to over 1,000 computer science students around the world. You can continue your lessons with his online demos and learning labs of the company’s AI technology on his website at the company.
Jeff Boudier, Head of Product and Growth at the company, said:
Hiring fresh out of college goes a long way.
Traditionally, the first day of class marks the beginning of the hiring cycle. However, they seem to be moving up the timeline as the competition heats up.
Nithya Attaluri, a Master of Engineering student at the Massachusetts Institute of Technology, said she started getting calls from recruiters early in the fall semester. A few months later, she got a job at her DeepMind (a subsidiary of Alphabet, Google’s parent company). And more Cornell seniors are coming back to campus from summer internships and getting offers to work full-time in the spring, Barra said.
For the few people who decide to pursue a PhD in academia, their expertise makes them the ultimate prize for headhunters.
Killian Weinberger, professor of computer science at Cornell University, said: “Most of them resist the temptation.”
“Most Stanford PhDs have startups or companies trying to get them to drop out or co-found,” said founder and CEO of Moonhub, a startup focused on automated hiring. Nancy Xu, Principal Investigator, said: She is currently on vacation at Stanford completing her AI PhD. A program focused on Moonhub.
Some of these companies are so keen to recruit graduate students that they allow them to work part-time while continuing their studies, said part-time study at call center automation startup Asapp. Weinberger, who also works as a Carnegie Michael Shamos, director of the Master’s Program in AI and Innovation at Mellon University, said he saw such part-time offers exceed $350,000 in base salary.
And full-time bids can be even higher. Carnegie Ilyas Bankole-Hameed, who is a student in his AI master’s program at Mellon University, told Insider that he’s heard of full-time offers amounting to nearly $500,000.
These staggering numbers are enough to lure many new graduates.In 2011, new AI Ph.D. graduates landed jobs in roughly equal proportions in the tech industry and academia. But since then, the majority of new graduates have gone to his AI industry, and the proportion of AI PhDs has nearly doubled. Will 2021 graduates find industry jobs or academic roles, according to his 2023 AI Index report from Stanford University’s Human-Centric AI Institute?
But salary is just one of the reasons for abandoning school for industry. University professors and students say the biggest draw is the opportunity to work on research that has real-world applications. And funding, often a limiting factor in academia, is plentiful in the startup space, with venture capitalists and tech giants like Alphabet and Microsoft funding small businesses.
“Every AI company has a role, whether they have a PhD or not,” said Attaluri, a future DeepMind researcher. It’s up to students to decide “how they want to spend their next five years,” she said.
unusual pool
The mad scramble for AI talent is driven in part by corporate leaders’ fear of falling behind, rather than by encouraging new technological advancements. His Shamos from Carnegie Mellon University told Insider:
“This is moving to the C suite, and corporate CEOs are starting to worry. What does this generative AI mean for us? And are we falling behind? If so, what can we do?” said Shamos.
Tribe AI, a job marketplace that connects companies with machine learning professionals, told Insider that the number of companies seeking help with data and AI projects doubled between December and March.
Glen Evans, core talent partner at venture firm Greylock, has seen a growing demand for expertise in this area, even among early-stage startups in the company’s portfolio. Applied research scientists are in particular demand, he wrote in an email. “Because of the industry’s focus on the larger language model and its widespread adoption across the board.”
Companies’ hiring needs depend on their technical approach, whether graduate students’ knowledge and skills are required for payroll, or just undergraduates.
There are two camps in the AI startup space. Many such as Jasper and Copy.ai build their apps on top of third-party large-scale language models, such as OpenAI’s impressive GPT-4. Others, such as Adept and Character.AI, are taking a “full-stack” approach, building everything from basic foundational models to the final product.
Creating these highly complex tech stacks requires highly technical AI talent.
“Someone can get this kind of understanding without a degree,” said Bankole-Hameed. “But if someone who has never been to medical school does the surgery, the doctor will not be trusted.”
And while these engineers and researchers are rarer and more expensive, they’re a more defensible approach that’s harder for competitors to replicate, VC told Insider.
According to Greylock partner Saam Motamedi, the number of computer scientists who can build and train their own models is very small: dozens instead of hundreds.
Madrona VC Jon Turow agrees.
tough decision
Job prospects are not yet attracting enough people to the field. Nearly two-thirds of his companies across nine industries, including banking, insurance, government and retail, said their biggest skills gaps were in AI and ML, according to SAS research.
Meanwhile, some universities are beefing up their curricula to prepare students with only bachelor’s degrees to work in this burgeoning environment.
Last fall, MIT added majors in artificial intelligence and decision-making. Harvard has opened a new research institute to study artificial intelligence. The lab was funded by donations from Meta founder Mark Zuckerberg and his wife, Priscilla Chan. Cornell is also developing two of his minors for those interested in AI.
Kevin Liu, an undergraduate student at Stanford University studying computer science and AI, said many of his classmates had taken a new interest in the field, forming new entrepreneurial clubs and having Telegram chats. said he saw them flocking to the school’s popular natural language processing course.
These opportunities for undergraduates could help reduce the labor shortage. Reed Simmons, a research professor at Carnegie Mellon University, says when he helped create an artificial intelligence program for undergraduates four years ago, most jobs only offered master’s or doctoral degrees. student. Recruiters confirmed their commitment to the undergraduate curriculum, so “many of these opportunities are open to undergraduates,” he said.
But there are still students like Eyzaguirre who continue their studies in academia despite the attractive salaries and benefits in front of them.
He is still in his second year of his doctoral program. After graduating, he hasn’t decided whether to continue his studies at university or join a fast-growing technology company. He acknowledges that fewer students are entering graduate school, and with the associated difficulties, such as the slow pace of research, heavy workloads, and disastrous wages for lab work, making a lot of money.
“We can do interesting research,” Eyzaguirre said of academia.
