UCSD - University of California - San Diego

08/31/2026 | Press release | Distributed by Public on 08/31/2026 10:41

A Summer of Learning by Doing

Published Date

August 31, 2026

Article Content

This summer, three groups of interns at the University of California San Diego Qualcomm Institute (QI) learned by doing - fabricating semiconductor devices, building serious games and developing artificial intelligence applications.

"There's no substitute for hands-on experience when it comes to building workforce-ready skills," said QI Director Ramesh Rao. "Our summer interns don't just learn about technology - they conduct research, write programs, use advanced equipment and build prototypes. They leave with practical skills they can carry directly into their next classroom, internship or job."

Small chips, big opportunities

New this year was a program providing community college students an opportunity to learn skills related to semiconductor manufacturing and nanotechnology.

Called Experiential Semiconductor Training in Emerging Nano-Technologies (ExSTENT), the nine-week internship, supported by a grant from the National Science Foundation, provided students with hands-on training in semiconductor fabrication techniques - including lithography, deposition and etching - using QI's Nano3 cleanroom and Makerspace. Interns also attended weekly career-readiness seminars with UC San Diego Center for Research on Education, Assessment and Teaching Excellence (CREATE).

In the ExSTENT showcase event on August 21, teams of interns competed for the honor of best poster. (Photo by Kristina Mallari/UC San Diego Qualcomm Institute)

"We had 160 applications for 20 spots," said Riley Need, a QI research specialist who leads the program with Oscar Vazquez Mena, professor in the Aiiso Yufeng Li Family Department of Chemical and Nano Engineering at the Jacobs School of Engineering. "These students are exceptional, and it was incredibly rewarding watching them gain confidence and learn skills that are in-demand by industry."

In the ExSTENT showcase event on August 21, four teams of interns presented posters highlighting their work on biotechnology, solar energy, photonics and quantum materials, with each team presenting on one of the four projects completed by all participants. For the biotechnology project, interns built a simple, low-cost test strip - similar to an at-home COVID test - to detect specific DNA or biological markers. For the solar project, they built and tested their own solar cells made from a light-absorbing material called perovskite - then traced disappointing results back to leaks in the manufacturing process, gaining firsthand experience with the troubleshooting required in a research lab. For the photonics project, they etched tiny optical circuits onto silicon wafers, essentially miniature pathways for light instead of electricity, and measured how well those circuits performed.

For the quantum project, which was awarded best poster, interns grew and tested ultra-thin films of niobium, the metal used in the superconducting circuits at the heart of quantum computers.

"Learning the actual manufacturing process behind the chips was immensely related to the field I want to go into," says intern Eitan H. (right), here receiving a certificate from Riley Need, ExTENT director and QI research specialist. (Photo by Kristina Mallari/UC San Diego Qualcomm Institute)

"Being able to get hands-on in the cleanroom and visit [Quantum Design] was really special," said Eitan H., a MiraCosta College student on the quantum team. "Learning the actual manufacturing process behind the chips was immensely related to the field I want to go into."

"As a community college student, it's very hard to get a look into industry," added teammate Nadav H. "The ExSTENT program gave us that close-up view, while also giving us a sense of what it will be like to be a student at UC San Diego when Eitan and I transfer here in the fall."

Serious games, real-world skills

A second internship track, NeuroDiversity in Tech (ND Tech), a program now in its ninth year, gave participants - many on the autism spectrum - hands-on experience building video games. A showcase event on August 14 brought together proud parents, mentors and friends to learn more about the interns' work.

"It's so great to see all the new dear friends and family, fantastic interns that we have for summer of 2026," said Shana Cohen, an autism researcher and associate professor in UC San Diego's Education Studies Department, who co-directs the ND Tech program with Neil Smith, QI associate research scientist and co-director of the UC San Diego Center for Cyber-Archaeology and Sustainability. "Why is this internship important? We have a growing population of neurodivergent folks coming into the workforce, and the unemployment rate is quite high across this group. There's also a need in this group for durable and soft skills. We spend a lot of time training people how to communicate effectively."

During the eight weeks of the internship, participants worked with each other - as well as clients, coaches and industry mentors - to build what organizers call "serious games": games meant not just to entertain, but to simulate, train or teach something real.

Intern Brandon Kearney (center) demonstrates "Hoyo Negro," a serious game the ND Tech students developed over the summer. (Photo by Mika Ono/UC San Diego Qualcomm Institute)

This year's interns developed two games started by last year's ND Tech cohort, giving the 2026 interns a chance to pick up an existing codebase, fix known issues and add new features - a common task in a real software job.

Roy Aizenberg, Anshool Yervia and Eduard Maranvilla presented their work that built on the inherited codebase for "Incident Commander," a top-down strategy game that puts the player in charge of directing firefighting squads through burning buildings to rescue civilians. The team added a scoring and ranking system, gave individual rescue units distinct behaviors, and used real geographic data to generate new levels.

Interns learned Unreal Engine's "blueprints" system and wrestled with the kinds of problems common to professional software development - including the complexity of a large game engine and conflicts that arose when multiple people edited the same files.

"This project made me realize that I learn things really fast," said Yervia. "At the end, I was pretty good at using Unreal Engine. My technical skills grew from this internship, and it further increased my interest in programming."

Intern Roy Aizenberg (right) talks a family member through the students' work. (Photo by Mika Ono/UC San Diego Qualcomm Institute)

Brandon Kearney, Sriram Venugopal, Wayne Su and JP Tasto worked with client Dominique Rissolo, QI research scientist with the Cultural Heritage Engineering Initiative, on "Hoyo Negro: Subterranean Wayfinding," a scientific simulation that reconstructs a submerged cave in Mexico as it existed before it flooded at the end of the last Ice Age. The cave, once dry, trapped and preserved the remains of more than 20 species of megafauna - including saber-toothed cats, ground sloths and mastodons - offering researchers an unusually complete fossil record.

Building on 3D scans of the real cave, the team improved the game's water-level controls and created a heat map tracking where virtual animals spent time, which could help researchers identify routes the real animals may have taken. Beyond the coding itself, the Hoyo Negro team said they came away with a better handle on real-world project management - Kearney pointed to using tools like Jira, Git and Google Slides to keep the team organized and track individual tasks.

Throughout the program, which was supported by the California Department of Rehabilitation and the Motorola Solutions Foundation, interns also attended weekly seminars with guest speakers from industry and workshops on professional skills like resume writing and LinkedIn profiles.

From algorithms to applications

The third track brought undergraduates from South Korea to QI for the Artificial Intelligence (AI) Development Project, an intensive program that pairs Korean students with hands-on training in machine learning, data analysis and entrepreneurship. Founded in 2013, the program runs sessions twice a year - one in the winter, focused on AI research, and one in the summer, focused on AI entrepreneurship - and has drawn students from many different South Korean universities. This year's session included 34 undergraduates from Keimyung, Kyonggi and Chosun Universities, as well as Kumoh National Institute of Technology.

The program is led by Seokheon "Justin" Cho, QI's principal network architect and a professor in South Korea. Rather than emphasizing theory, it's built around a single structured project: over four weeks, teams move from identifying a real-world problem to data analysis, AI model development and cloud deployment - using tools including Google Cloud Platform, GitHub and Claude Code - before presenting a finished, working platform.

Intern Seungchae Lee presents his team's AI entrepreneurship project, "D-avocado," an app that predicts an avocado's ripeness from an image. (Photo by Matthew Morris/UC San Diego Qualcomm Institute)

This year's seven teams built a wide range of working AI prototypes. One team - which won first prize at the showcase event on July 24 - built an app designed to classify skin lesions from a smartphone photo and suggest a next step, such as consulting a nearby dermatologist or following a home-care guide. An app that predicts an avocado's ripeness from a photo took second. A tool that calculates a vehicle's aerodynamic drag coefficient from a 3D model took third. A separate scam-detection and speaker-analysis tool also earned outside recognition in the ML Empowerment Build Challenge 2.0.

Building the prototypes required students to move beyond classroom exercises, making decisions about everything from training data and model selection to user experience and how to communicate an application's value. In addition, the interns learned how to tell the story of their technology as an entrepreneur.

"You can learn the theory and algorithms in a classroom, but building an actual AI application forces you to think differently," said Cho. "Students have to decide what problem they're solving, choose the right data and methods, troubleshoot when things don't work as expected, and explain why their solution is useful. That process turns knowledge into practical experience."

Whether they were working in a cleanroom, debugging a video game or training an AI model, QI interns across programs faced a common challenge: taking what they knew and making something work. Along the way, they practiced not only technical skills but also troubleshooting, teamwork, communication and project management - skills that are difficult to develop through classroom learning alone.

To learn more about QI's educational programs and internships, see QI's Education web pages.

Learn more about research and education at UC San Diego in: Artificial Intelligence

Read more news about: School of Social Sciences, Artificial Intelligence

UCSD - University of California - San Diego published this content on August 31, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on August 31, 2026 at 16:41 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]