UC Davis Engineering Student Wins International Prize for AI-Powered Harvest Technology
Luis Blas, a fourth-year electrical and computer engineering undergraduate at the University of California, Davis, has earned a top prize at an international invitation-only SD Express Student Project Competition for developing HarvestIQ, his agriculture-focused edge AI platform.
The SD Association, or SDA, is a global, industry-led organization that develops and promotes SD memory card storage standards. The association launched the SD Express Student Project Competition in 2025 and invites participation from highly ranked universities worldwide, targeting the top 150 institutions of the Academic Ranking of World Universities.
The 20 students from six universities who competed this year received an NVIDIA Jetson Orin Nano Super Developer Kit, a powerful, compact generative AI computer, equipped with a microSD Express memory card. The students could apply the technology however they saw fit; the only requirement was to demonstrate how the card was used to implement the submission.
A Tool for a Smarter Harvest
Blas’s HarvestIQ, an edge AI harvest station that assesses crop quality in the field in real time, tied for third place in this year’s competition.
In agriculture, quality checks in the field are mostly visual and depend heavily on each grower’s judgment. That information rarely reaches packing centers, so when fruit gets damaged during transport, growers are often blamed for problems caused by outside factors.
HarvestIQ uses a camera and a smart scale to capture each harvested fruit — for this project, Blas focused on strawberries. Lightweight AI models running on the Nvidia computer check for ripeness as well as visible damage or disease. Concurrently, every image, weight and prediction is logged to the microSD Express card, allowing the system to work fully offline.
The card also stores the trained models and system settings, so growers can transfer HarvestIQ’s capabilities to another device simply by swapping the card. Users can also correct the system’s predictions, and those corrections are saved to improve future models. Blas tested the system in local strawberry fields with the help of area growers.
Best Fruit Forward
With this technology, Blas aims to give growers an accurate, consistent record of fruit quality at harvest, building greater transparency into the process and reducing the quality-control burden on people in the field. Blas, who was raised in the Salinas Valley and Watsonville areas of California, designed HarvestIQ with those farming communities in mind.
“This competition gave me a chance to share what I care about: building technology for agricultural and environmental settings around human-centered design,” he said. “I grew up in a farming community with a family that works in this industry. They’re the reason I built something like HarvestIQ, and it means a lot to have the recognition.”
Blas, who was advised on this project by Professors of Electrical and Computer Engineering Weijian Yang and Zhi Ding, will continue developing HarvestIQ while maintaining his human-centered design focus.
“Agtech is still an emerging field. There is still a lot of room to grow,” Blas said. “My goal is to expand HarvestIQ to other crops and eventually into in-field robots and wearable assistants — technology that collaborates with farmworkers and with people in general. Building something for my community is what keeps me going.”
