Aug 2021 – Present · 4 yrs 7 mos · Ann Arbor, Michigan, United States

• Worked on problems related to motion understanding using self-supervised learning, video models and multimodal learning, published papers at CVPR 2023, ECCV 2024, CVPR 2025.

May 2021 – Aug 2021 · 3 mos · Sunnyvale, California, United States

• Worked on a new benchmark released for human–human, interactive dialogues to complete household tasks in simulation on AI2-THOR, published at AAAI 2022.

May 2020 – Aug 2020 · 3 mos · Sunnyvale, California, United States

• Worked on building agents that can solve a navigation task by holding dialog with each other in a cooperative setting, published at ACL 2022.

Aug 2019 – Apr 2021 · 1 yr 8 mos · Atlanta

• Worked on problems that related to Embodied AI and Vision & Language.
• Papers published at ECCV 2020 and CoRL 2020.
• Organized VQA Challenge 2020, 2021 and VQA and Dialog Workshop at CVPR 2020, CVPR 2021.
• Co-organized Visual Dialog Challenge 2020.

Aug 2018 – May 2019 · 9 mos · Atlanta

• Worked on solving navigation instruction following in 3D environments by following an ideal agent trajectory. Presented at NeurIPS 2019. Supervised by Prof. Devi Parikh and Prof. Dhruv Batra.

Apr 2018 – Aug 2018 · 4 mos

• Mentored a student for Fabrik project in CloudCV organization. Added support for importing/exporting models from TensorFlow. Built real time collaboration feature where multiple users can edit or review the model at the same time.

May 2017 – Jul 2017 · 2 mos · Singapore

• Built fast semantic segmentation models for autonomous driving by reducing complexities in model architecture. Explored and combined different deep learning models like PSPNet and MobileNets.

May 2016 – Jul 2016 · 2 mos · New Delhi, India

• Implementation and performance analysis of Extreme Learning Machines and its variants on object detection and blind blur detection.