May 2024 – Dec 2024 · 7 mos · Seattle, Washington, United States · On-site

• Built a method to adapt multi-modal large language models using online reinforcement learning to train embodied agents that can interact with the environment and communicate with humans in natural language to resolve ambiguity when required.

May 2023 – Aug 2023 · 3 mos · Seattle, Washington, United States · On-site

• Built a method leveraging off-the-shelf vision foundation models to learn visual common sense reasoning for localizing semantically meaningful placement for objects.

May 2022 – Aug 2022 · 3 mos · Boston, Massachusetts, United States

• Worked on building embodied agents that can learn to navigate and interact with objects using 3D scene graphs.

Aug 2021 – Present · 4 yrs 7 mos · Atlanta, Georgia, United States

• My research focuses on building agents that can see, reason, and interact reasonably. Currently, I am working on building embodied agents that can learn to explore from humans by imitating human-like behavior when trying to solve a task.
• I am advised by Prof. Dhruv Batra and Abhishek Das and I collaborate closely with Prof. Devi Parikh.

Feb 2020 – Aug 2020 · 6 mos

• Working as mentor for EvalAI project at CloudCV. Guiding contributors participating in Google Summer of Code 2019 to understand how open source development works.
• Responsibilities involved are:
• Reviewing work submitted
• Helping them understand the project and resolve any difficulties faced.
• Review the applicant proposal for a project
• Help students throughout the GSOC timeline to complete the deliverables of the project.

Dec 2019 – Jan 2020 · 1 mo

• Working as a mentor for EvalAI project and managing other mentors at CloudCV. Guiding pre-university students to understand how open source development works.
• Responsibilities involved are:
• Reviewing work submitted
• Helping them understand the project and resolve any difficulties faced.

Aug 2019 – Aug 2021 · 2 yrs · Bengaluru, Karnataka, India

• My work is around developing features that increase engagement of the platform using multiple software development frameworks and my knowledge of machine learning.

Jul 2019 – Aug 2019 · 1 mo

• Working on implementing ETL pipelines to process large scale analytics data on daily basis and implementing cache infrastructure support to improve the stability of our system.

Jun 2018 – Jun 2019 · 1 yr

• Working on implementing targeted content delivery system, content personalisation system using machine learning techniques and data pipelines for processing large scale analytics data.

Apr 2018 – Aug 2018 · 4 mos · Pune Area, India

• Worked with CloudCV to make designing deep learning models easier with the help of Fabrik. Fabrik is an online collaborative platform to build, visualize and train deep learning models via a simple drag-and-drop interface. It allows researchers to collaboratively develop and debug models using a web GUI that supports importing, editing and exporting networks written in widely popular frameworks like Caffe, Keras, and TensorFlow.

Jan 2018 – Jun 2018 · 5 mos · Pune Area, India

• Worked with InterviewBit for designing a platform to help users learn skills needed for technology jobs by adding end to end support for new features in InterviewBit website.

Oct 2017 – Mar 2018 · 5 mos · Pune Area, India

• Worked on designing algorithmic problems for competitive coding contests.

Jul 2017 – May 2018 · 10 mos · Pune Area, India

• Worked with NVIDIA's multimedia team for developing a test suite for NVIDIA's video encoder, providing highly efficient test suite execution environment which will allow developers to test the video encoder. And increasing the coverage of the tool which automates the tests by adding support for multiple category of tests for multimedia encoders and decoders.