I am a first year Ph.D. student in Robotics at the University of Michigan, Ann Arbor, advised by Professor Bernadette Bucher in the Mapping and Motion Lab. My research sits at the intersection of robotics, 3D perception, and natural language processing. Specifically, I study how representations of real-world environments influence decision making in large vision-language models performing exploratory embodied tasks such as interactive object search.

Prior to my graduate studies at the University of Michigan, I was a Software Development Engineer at Amazon Fulfillment Technologies and Robotics, where I designed large-scale task assignment and motion planning solutions for the 1,000,000+ mobile robots in Amazon’s fulfillment network. I earned my Bachelor of Science degree from Cornell University in 2022, double majoring with honors in Electrical and Computer Engineering and Computer Science.

Publications

CVPR 2026 | Workshop on Multiagent Embodied Intelligent Systems

PhyGS: Physically Grounded Controllable Scene Generation

Aparajito Saha, Zhen Hao Gan, Jinjia Guo, Jacob Skwirsk, Jeremy Acheampong, Anton Arapin, Chahyon Ku, Yue Hu, Nima Fazeli, Bernadette Bucher

Projects

ForageBench: A Photorealistic, Physically Grounded Benchmark for Interactive Object Search

Aparajito Saha, Zhen Hao Gan, Jinjia Guo, Jacob Skwirsk, Jeremy Acheampong, Anton Arapin, Chahyon Ku, Yue Hu, Nima Fazeli, Bernadette Bucher

Teaching

EECS 442: Computer Vision

Graduate Student Instructor, University of Michigan, Department of Electrical and Computer Engineering, 2025

EECS 442 is an introductory course in computer vision, covering topics such as low-level vision, 2D signal processing, object recognition, image synthesis, 3D reconstruction, and deep learning.

Teaching Materials

For full course details, please visit the course website.

ECE 5725: Design with Embedded Operating Systems

Teaching Assistant, Cornell University, School of Electrical and Computer Engineering, 2021

ECE 5725 is an advanced course focused on the design of microcontroller systems using embedded Linux. Student teams design and debug projects on a Raspberry Pi, with emphasis on application and Linux programming alongside control of external hardware.

For full course details, please visit the course website.