The Cal Poly AI Factory opened on October 22. The California university invested US$ 3 million in a facility with four NVIDIA supercomputers, according to Mustang News, the student-run news outlet at Cal Poly.
In practice, students and faculty from every college at the school will be able to build and run AI models right on campus. It is a real-world example of training AI professionals with a hands-on focus. It also offers useful clues for Brazilian companies that need to get their teams ready for the technology.
The launch was part of the first AI Day at the university, a full-day event with talks, a research showcase and a digital countdown to switch on the new facility. Cal Poly President Jeffrey Armstrong attended, along with Charlie Wuischpard, NVIDIA vice president for the Americas. According to the report, Armstrong also thanked Rep. Jimmy Panetta, who presented the university with a Congressional recognition for creating the space.
How access will work
The funding came from the Noyce School of Applied Computing, the applied computing school at Cal Poly. The machines are housed in the Frank E. Pilling building on campus. A university statement cited by Mustang News says only a few universities in the United States offer this kind of machine learning resource.
Access will not be automatic. As spokesperson Matt Lazier explained to the outlet, anyone who wants to use the facility will log in through the university portal after finishing a free training course. The IT team is still setting up that course, so full access depends on this step.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
The projects shown at the event give a good sense of how the facility will be used. Students in the Business Analytics master program built forecasting models with more than 200 million rows of data, according to the report. They worked with Jess Poirier, an NVIDIA data scientist and Cal Poly alum. Bennan Davis, the program director, told Mustang News that the dataset is too big to handle without help from the campus facility, the faculty and NVIDIA itself.
There were also projects focused on agriculture and business. A statistics student, James Riggs, told the outlet he came to the event with his career in mind. He especially liked a multi-agent AI system built to support engineering students.
The NVIDIA message about work and AI
In his keynote, Wuischpard talked about the growing use of AI in robotics, healthcare and agriculture. He also noted that demand for computing power usually outpaces what is available, which is why an on-campus facility helps academic research in the region.
But the part that drew the most attention was about jobs. According to Mustang News, the executive called AI the biggest transformation we will live through. He said the risk is not losing your job to an AI, but to someone who knows how to use the tool better.
He also praised the Learn by Doing model at the university, which is built around learning through practice. In the view of Wuischpard, as the report notes, tens of thousands of companies could benefit from the practical skills these students are building.
Armstrong, for his part, acknowledged that people on campus have mixed views on AI. Even so, he argued that it is necessary to understand the technology and to train people who can set rules and limits so it is used in a positive way.
What this means for people working with AI in Brazil
Cal Poly is far from here. Still, the way it set up this facility offers a few lessons that fit the daily routine of Brazilian companies. What follows is analysis by Método Viral, not information from the report.
The first lesson is that artificial intelligence training tends to pay off more when it starts from real data and real problems. The students did not stop at theory. They took a huge dataset and went looking for answers. Picture an online store that, instead of a generic course, trains its team using its own sales history. The odds that the learning makes sense in daily work tend to be higher. We have covered this before in how AI is changing the way teams get trained.
The second one is easy to copy. Requiring training before giving people access to tools helps prevent misuse and builds a shared knowledge base across the team. Assessment tools, like the one we showed in our piece on NVIDIA SkillEvaluator in action, can round out this process.
The third has to do with privacy. Sophie Kumar, a computer science student, told Mustang News that she worries about where the processing happens and does not want her data sent elsewhere. If students are already thinking about this, Brazilian companies have every reason to do the same, since the LGPD, the Brazilian data protection law, requires care with personal data.
Finally, the link between universities and artificial intelligence shown here points to a path for smaller businesses that lack the budget to build their own infrastructure. Partnerships with local colleges can open doors to research, talent and hands-on projects, much like what already happens in the world of partnerships that are changing startups.
A powerful machine, on its own, solves nothing. The Cal Poly case shows that the investment makes sense when it comes with people who are ready to use the technology well. And training AI professionals is still the part that carries the most weight.
