Mechanical Engineering student at Carnegie Mellon building systems that sit at the intersection of robotics, AI, and real-world impact.
Building intelligent robotic systems for product testing at the Chemical Analysis Division.
Investigated the biomechanical efectiveness of combined hip and knee exoskeletons to increase stability and correction upon perturbation detection, instability, and imbalance through user testing and simulation. Also designed rigid, adjustable exoskeleton systems around an NVIDIA Jetson to decrease sensor-reading fluctuation.
Focused on discovering the intersection of the physical world and AI. Primary coursework in AI/ML, computer vision, controls for robotics, and mechanical engineering principles.
Working through ambiguous, unstructured problems and molding them into structured, actionable solutions. Collaborating with high-performing teams to think critically and iterate quickly.
Contributed to a nationally competitive team. Led rapid design, build, and test cycles via CAD-driven prototyping. Implemented advanced control systems featuring object tracking, odometry-based localization, and FSMs.
I'm a third year Mechanical Engineering student at CMU focused on robot learning—bridging perception, control, and motion planning to enable machines to act effectively in the real world.
My work spans closed loop control systems, trajectory optimization, and learning based approaches to improve robot performance and intelligence. I'm drawn to problems where robust mechanical design meets intelligent algorithms: building systems that can adapt, learn, and execute with precision under real world constraints.
I approach engineering through rapid prototyping and iteration—from CAD and fabrication through control validation—reducing the gap between simulation and physical deployment.
A real-time hand gesture recognition system that enables touchless video playback control through machine learning.
An ankle-based perturbation controller to enable realistic testing of lower-limb exoskeletal systems under trip and slip conditions.
Design and manufacturing of weight-optimized mechanisms including a reverse four-bar linkage and a differential 3-axis wrist.
A GPT-style language model built from scratch with custom multi-head attention and trained on text data.
An implementation of the AlexNet architecture from scratch and fine-tuning experiments on ASL datasets.