Turning ideas into robots, and challenges into code

Passionate about robotics, I bring diverse experience across AMRs, cobots, quadrupeds, and advanced robotic systems. With a strong foundation in autonomy, perception, SLAM and Deep learning, I specialize in designing intelligent solutions that improve efficiency and expand real-world applications of robotics.

Skills

Technologies and tools I have mastered over the years


Python
C++
ROS2
ros
Gazebo
PyTorch
tensorrt
Rviz
isaacsim
jetson
moveit
git
linux

Linux

opencv
numpy
pcl
eigen

Eigen

solidworks
matlab
studio5000
jira

Projects

Unitree Go2

Developing navigation and mapping systems for a Unitree Go2 quadruped to operate effectively in constricted spaces

Person Re-Id using Siamese Networks

Development of a Siamese Neural Network for one shot learning

Attention Guided Off-Road Semantic Segmentation

Methods capable of accurately identifying and classifying various terrains, obstacles, and objects for improved autonomous vehicle navigation

Dead Reckoning

Estimating the trajectory of a vehicle using data from an IMU. Uses techniques like IMU calibration, complementary filters, and sensor fusion

Comparative Study of SLAM

Comparing SLAM algorithms - RTAB-Map, OpenVSLAM, Cartographer and SPTAM

3D Modeling

Check out some robots and mechanisms that I designed and prototyped.

Education

Northeastern University

Master of Science in Robotics

Northeastern University, Boston, MA

September 2022 - Dec 2024

GPA: 3.95/4.0

Relevant Courses:

  • Mobile Robotics
  • Deep Learning
  • Robotic Sensing and Navigation
  • Robot Mechanics and Control
  • Reinforcement Learning
  • Pattern Recognition and Computer Vision
  • Control Systems
KJSCE

Bachelor of Technology in Mechanical Engineering

K. J. Somaiya College of Engineering, Mumbai, India

Aug 2017 - May 2021

GPA: 8.84/10.00

Experience

Piaggio Fast Forward, Boston, MA

Role: Robotics Intern

Dates: July 2023 - April 2024

Responsibilities:

  • Implemented Iterative Closest Point (ICP) to enhance localization, resulting in a 35% improvement in the position estimate.
  • Developed a dataset creation tool in Python for leader identification using data collected from a person-following robot.
  • Trained a Siamese Neural Network for leader re-identification in a one-shot learning framework using PyTorch, achieving 90% accuracy, and efficiently deployed it on Jetson TX2 using the TensorRT C++ API for real-time inference.
  • Designed and implemented a data collection tool in C++ to capture LIDAR readings and ground truth data from a laser rangefinder, enabling the creation of a robust sensor model.
  • Engineered a LIDAR sensor model using linear regression on surface data, significantly improving accuracy for precision docking across varying surface types and optimizing real-world performance.
  • Developed precision docking feature for an autonomous mobile robot with onboard RP LIDAR and RealSense depth camera.

IITD-AIA Foundation for Smart Manufacturing, IIT Delhi, India

Role: Robotics Engineer

Dates: August 2021 - July 2022

Responsibilities:

  • Integrated an Autonomous Mobile Robot (AMR) with an onboard TM5M700 cobot for autonomous material handling in a Cyber-Physical factory using ROS.
  • Developed an optimized pick and place sequence for workpiece handling with cobot using the Moveit! ROS package.
  • Optimized workpiece detection by creating an object detection pipeline using transfer learning.
  • Deployed the workpiece detection pipeline on NVIDIA Jetson Xavier & merged it with the pick-and-place pipeline.
  • Defined waypoints for AMR to automate the AMR dispatch process for seamless operation using ROS and REST API.

Certifications

Resume

Download my full resume here.

Get In Touch

I'm always open to discussing opportunities, projects, and collaborations in Robotics. Feel free to reach out if you'd like to work together or learn more about my work.