Yash Kakade

Engineering Applied Science Junior at Caltech

Hey! I’m Yash, a third-year undergrad at Caltech studying Electrical Engineering with a focus on Intelligent Systems. I’m advised by Prof. Babak Hassibi.

Broadly, I’m interested in embodied intelligence at the intersection of hardware, perception, and learning-based control. My work spans robotics, computer vision, motion capture, and reinforcement learning, with a strong emphasis on real-world deployment. At Caltech, I work on humanoid locomotion using reinforcement learning in the AMBER Lab. In parallel, I conduct perception and robotics research with Relativity Space, focusing on high-precision sensing, motion capture, and digital twin systems for aerospace applications.

I’ve previously worked as a robotics and computer vision intern at AFRL developing ROS2-enabled digital twins and perception systems for autonomous robots, as well as a Robotics Software Engineering Intern at Relativity Space, and briefly conducted machine learning research in the Wierman Group at Caltech on adversarial robustness in CNN-based perception.

Yash Kakade

ykakade [at] caltech [dot] edu

Selected Projects

Autonomous Rubik's Cube Solving Robot

Autonomous Rubik's Cube Solving Robot

6-DOF Robot Arm2026

A 6-DOF robot arm that autonomously scans, solves, and executes moves on a physical Rubik's cube using computer vision and a custom manipulation pipeline.

Computer VisionManipulationState Machine
Camera-Only Motion Capture System

Camera-Only Motion Capture System

10 micron level accuracy2025

Self-healing calibration and rigid-body tracking system requiring only standard cameras.

Computer VisionPerceptionMotion Capture
Humanoid Tennis Manipulation

Humanoid Tennis Manipulation

5D|Redundancy-Aware IK2025

Autonomous tennis system with a 29-DOF humanoid using hybrid IK and whole-body control.

ROS 2Inverse KinematicsDamped Least SquaresCode
Universal Object Tracking & Digitization

Universal Object Tracking & Digitization

92.8% Detection Accuracy2024

Automated digital twin pipeline with high-accuracy object detection for Boston Dynamics Spot.

Digital TwinComputer VisionSimulation
Fully Autonomous Low-Cost Lawnmower

Fully Autonomous Low-Cost Lawnmower

<$300 Build vs $4k Commercial2022

A sub-$300 award-winning autonomous lawnmower capable of 2D SLAM and obstacle avoidance.

SLAMNavigationHardware

Current Focus

Balancing engineering, research, and studies at Caltech.

Relativity Space

Robotics Software Engineering Intern

Relativity Space
PerceptionMotion CaptureStructure from Motion

Caltech

Undergraduate Researcher

AMBER Lab
ControlReinforcement LearningHumanoid Locomotion
June 2027Graduation
Caltech

California Institute of Technology

GPA: 3.7/4.0

B.S. Electrical Engineering (Intelligent Systems) and Business, Economics & Management (BEM)