cv

Applied AI / ML Software Engineering Intern

Brain Corporation

Contributed to AI powered retail automation systems for shelf inventory monitoring.

• Trained detection models on Google Vertex AI that identified out of stock items with high precision.
• Curated and annotated a custom image dataset using V7 Darwin and BigQuery.
• Deployed optimized models into production pipelines improving inventory checking speed.

AI and Robotics Engineer

Argon Robotics

Developing perception and control software for fully autonomous welding robots in the construction industry.

• Created 3D reconstruction pipelines that improved weld seam localization accuracy.
• Integrated ROS2, MoveIt2, and xArm APIs for precise manipulator motion during welding tasks.
• Optimized control algorithms that reduced positional error and produced smoother arm trajectories.

AI and Computer Vision Engineer

NeuronFlo

Developed real time PPE compliance monitoring solutions for industrial sites.

• Trained PyTorch object detection models that achieved reliable helmet and vest recognition.
• Ported inference to NVIDIA Jetson Nano enabling onsite deployment with limited power budgets.
• Implemented alert pipelines that improved workplace safety through timely notifications.

Software Engineering Lead

Innovative Marine Technology Lab, Scripps Institution of Oceanography

Leading software development for autonomous underwater robots used in oceanographic research.

• Built sensor and camera pipelines that ensured robust data capture in harsh marine environments.
• Designed a web based interface that streams live data and allows remote control of onboard devices.
• Created computer vision tools to quantify underwater bubble dynamics for climate studies.

Software Team Lead

Yonder Deep

Led software efforts for low cost autonomous underwater vehicles used in Arctic research.

• Built Python control software that integrated sensor data with navigation routines.
• Achieved the first fully autonomous dive of the student built robot.

Technical Chair and Mentor

IEEE at UC San Diego

Ran workshops on computer vision and deep learning and mentored competition teams.

• Taught OpenCV and neural network fundamentals to more than thirty students.
• Guided the Signal Processing Cup team to achieve ninety five percent accuracy in Deepfake detection.

Outreach Chair

Eta Kappa Nu (HKN), UC San Diego

Led K-12 engineering outreach and created hands on lessons in electronics and coding.

• Organized classroom visits that reached more than one hundred fifty students.
• Developed a new ECE 198 course that trains undergraduates in outreach pedagogy.

Master of Science in Electrical and Computer Engineering

University of California San Diego

Intelligent Systems Robotics and Controls

• Stochastic Dynamical Systems
• Convex Optimization
• Robot Learning
• Deep Learning
• Planning and Learning in Robotics
• Robot Manipulation and Control

Bachelor of Science

University of California San Diego

Electrical Engineering, Mathematics Computer Science, and Physics

• Machine Learning
• Computer Vision
• Robotics
• Optimization

Apple Pathways Scholars Award

Apple Inc.

Fifteen thousand dollars per year merit scholarship recognizing academic achievement and leadership in inclusion initiatives.

UCSD ECE Best Tutor Award

UC San Diego Department of Electrical and Computer Engineering

Honored for outstanding teaching contributions across more than twenty courses.

Most Innovative Award, IEEE F1TENTH Competition

IEEE Intelligent Vehicles Symposium

Recognized for a novel deep learning based navigation system for autonomous racing.

Open Vocabulary Semantic Part Segmentation of 3D Human

International Conference on 3D Vision

Introduces an open vocabulary approach for part segmentation of human meshes and demonstrates strong performance on in the wild datasets.

Visual Inertial SLAM for Localization

Built a Python EKF based system that fuses stereo vision and IMU data achieving real time trajectory and landmark estimation.

• Downsampled features and used sparse matrices to meet real time constraints.
• Validated on multiple driving datasets showing accurate mapping.

FutureGAN for Human Motion Prediction

Implemented a generative adversarial model to predict future video frames of athletic movements using optical flow guidance.

• Trained on HMDB fencing sequences and achieved realistic frame generation.
• Evaluated with FID and Inception metrics to ensure quality.

Autonomous Navigation for F1TENTH Racecar

Led a team to deploy a CNN based controller that predicts steering and throttle from RGBD input earning fourth place in competition.

• Optimized inference on NVIDIA Jetson hardware using ROS2 nodes.
• Won the competition innovation award for creative design.

Quadcopter Flight Control and PCB Design

Designed a four layer PCB and embedded PID controller in C to stabilize a custom quadcopter platform.

• Integrated Atmega128RFA1 microcontroller and secure RF link.
• Achieved stable flight through sensor fusion of gyro and accelerometer data.