top of page
OVERVIEW
INTEL CORPORATION stands as a pioneering technology company on a global scale, fostering connectivity and innovation across borders. Driving innovation and connectivity across the world,
with a steadfast commitment to providing cutting-edge solutions, Intel Corporation stands as a cornerstone in the tech industry. In my role as grade level-7 user experience designer, I contributed to building the internal design system and worked as the lead designer on Intel’s new software platform for building computer vision models from video clips This effort will be migrated into Intel® Geti™ in the foreseeable future. Intel® Geti™ streamlines resource-intensive activities such as data uploading, labeling, training, model optimization, and retraining.
DISCLAIMER: The visuals and content have undergone modifications. The presented contents serve as a showcase of my design thinking practice and applied methodologies.
PROJECT ANALYSIS
CHALLENGE: As Intel® Geti™ was in development, it's program counterpart had been in the backlog for 8 years. As the production was in the rollout, I was assigned to update the designs of the video program to bring it into compliance with the new design system that my team and I had recently created. I also needed to create a "light mode" design that had not yet been considered. The objective was to create an intuitive interface that would empower users to utilize the program proficiently without the requirements of extensive training or costly educational resources. This demanded a meticulous design approach focused on user-friendliness and clarity, ultimately
resulting in a seamless experience for all levels of users.
DESIGN SOLUTION: Collaborating closely with stakeholders, we strategized to align the video
AI model training program closely to the image AI training model, in anticipation of their eventual integration. Despite the substantial similarities between these programs, it was imperative to acknowledge and address the nuanced disparities inherent to video and image models. This involved implementing techniques like creating snippets and snippet stacks, as well as employing a probability plot for model training. Over the course of six months, we successfully transitioned an eight-year-
old project into a fully operational state. Notably, our efforts were not solely focused on technical proficiency, but also on optimizing user-friendliness to its highest potential. This comprehensive approach ensured that the final product not only met its technical requirements but also provided
an intuitive and accessible experience for end-users of all proficiency levels.
TARGET USERS:
• Novice users (various backgrounds + industries)
• Engineers
• Manufacturing
• Retail
• Smart Cities
• Research
• Healthcare
• Smart Agriculture
DELIVERABLES: User flows, High-Fidelity Prototypes + Wireframes
TOOLS: FIGMA, FIGJAM, AI, PS
DURATION: 6 MONTHS
ROLE: LEAD DESIGNER
TAGS: #UX #UI #AI #SAAS
IoT/EDGE DEVELOPER JOURNEY MAP
To better understand the needs of our target users, I interviewed many people within my Intel network who work as IoT/edge developers. These interviews provided insights into their day-to-day work and approach to projects, which I used as a baseline to simplify the platform's user flow for more novice users while still supporting developer needs.
USER PERSONA + USER FLOWS
After gaining an understanding of the day-in-a-life of an IoT/Edge developer, I created two user persona profiles representing our key target audiences. To complement these, I developed workflows showing how users would create annotations and interact with the overall interface. My goal was to improve the usability and experience for both novice and expert users based on the research I conducted.
WIREFRAMES + EXPLANATION
COMPUTER VISION AI MODELS: By simplifying labor-intensive data upload, labeling, training, model optimization, and retraining tasks, teams quickly build vision models for a wide range of processes, including detecting defective parts in a production line, reducing downtime on the factory floor, automating inventory management, or other digitization and automation projects. The Intel® Geti™ software platform makes it intuitive and shortens the time-to-value of AI model development dramatically.
+ Sketchbook Designs
+ Intel® Geti™
APPLICATION HIGHLIGHTS
High-fidelity mockups and video walk-through of the application experience.
REFLECTION + RESULTS
The process of designing Intel Geti was a journey marked by meticulous planning, iterative development, and a profound commitment to user-centricity. Our team worked in close collaboration with stakeholders to ensure that the platform catered to the diverse needs of both novice users and seasoned experts. Striking the right balance between technical sophistication and user-friendliness proved to be a significant challenge. However, through a combination of intuitive interfaces, step-by-step guidance, and advanced features, we were able to create a powerful tool that empowers users at all proficiency levels.
Intel® Geti™ was successfully launched in Spring 2023 with updates for video annotation to come at a later launch date. The launch of Intel® Geti™ represented a culmination of months of dedicated effort and the realization of a vision. The platform's introduction to the market was met with enthusiastic reception, as it addressed a critical need in the AI model development landscape.
bottom of page