Keshav Mehndiratta

• Spearheaded perception development for two UAVs (fixed-wing and quadcopter) for the AEAC aerospace competition • Designed and implemented a YOLOv8 + ROS2 system on the fixed-wing drone to detect red balloons simulating wildfire origins; enabled autonomous navigation and IR-based fire mapping to a KML file • Built a precision landing system for the quadcopter using OpenCV and ROS2 for autonomous water collection and delivery • Mentored 15+ team members on ROS2, PX4, Gazebo, and perception pipelines, significantly improving onboarding speed and maintaining high code quality

• Modernized the user interface of ASSURIoT, an enterprise-level supply chain platform, by identifying and updating legacy Angular components to improve usability and maintainability • Developed and integrated new front-end features using Angular, working with existing backend REST APIs to enhance data visualization and streamline user workflows • Contributed to noticeable performance improvements and user engagement, including faster load times and increased user session durations over a 3-month period

• Identified the need for digitizing lab equipment management in undergraduate physics courses • Built a SQL-based backend and designed a responsive frontend using HTML, CSS, and JavaScript • Delivered a centralized equipment inventory portal now used by 200+ students and faculty per term

• Built parts for undergraduate projects in the machine shop using Miter Saw, Milling Machine and Lathe • Digitized lab and equipment manuals in the undergraduate physics labs for cloud storage • Tested the lab equipment and troubleshoot/identify any issues

First AI startup school. Selected as part of the top 2000 CS undergrads, grad students and recent graduates in AI to hear from the world’s top AI experts and founders in SF.