A mathematician and engineer with a passion for training neural networks. My journey spans across diverse interests from crafting algorithms and playing board games to solving puzzles, reading poetry, and nurturing plants.

Education

• Ph.D. in Robotics, (soon)
• MSc in Applied Mathematics, Northeastern University (Boston, USA)
• BTech in Mechanical Engineering, NMIMS (Mumbai, India)

My academic background covers a range of subjects, including mathematical modeling, optimization, machine learning, deep learning, and mechanical engineering principles.

Research Interests

My focus revolves around enhancing autonomous mobile robots’ capabilities through advancements in 3D environment modeling and navigation of unknown terrains. I leverage deep neural networks, graphical models, and optimization techniques to improve sensing and decision-making in these systems.

Experience

Machine Learning Engineer - DEKA Research and Development Corp.

• Autonomous Delivery Robot: Developing deep learning sensor fusion models for comprehensive global occupancy grids, integrating data from lidar, radar, and cameras.
• Insulin Pump Enhancement: Employing Reinforcement Learning and Transformer techniques to optimize insulin delivery without relying solely on user-input meal information.
• Infusion Pump Analysis: Utilizing vision detection algorithms to calculate liquid flow rates accurately.

Machine Learning and Statistics Research Assistant - Nano-medicine Center (NEU)

• Implemented automated classification models on MRI data, achieving an 83% correct classification rate for Alzheimer’s disease.
• Engineered segmentation models for MR images, extracting crucial gray matter characteristics.

Teaching Assistant and Grader - Northeastern University

• Mentored students in mathematical methods, modeling, and calculus courses.

Skills

• Programming Languages: Python, C++, MATLAB
• Software/Tools: PyTorch, CUDA, Pandas, NumPy, Scikit-Learn, OpenCV, ROS, Linux

Projects

I’ve undertaken various projects exploring attributes distributions, mathematical modeling of idea spread, statistical analysis of crime and weather correlations, probabilistic algorithms, and data-driven predictions. Some highlights include capset problem distribution exploration, idea spread dynamics, crime-weather associations, Sudoku solving algorithms, drug failure prediction, and personalized book recommendations.

I’m driven by curiosity and a fervor for leveraging mathematical and engineering concepts to solve real-world challenges. Let’s connect and explore the possibilities!