Feb 2023 – Apr 2025 · 2 yrs 2 mos · Sunnyvale, California, United States · On-site

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

• Worked with Trustworthy Alexa AI to explore the federated learning for NLP

Aug 2017 – Jan 2023 · 5 yrs 5 mos · Los Angeles Metropolitan Area

Jul 2016 – May 2017 · 10 mos · IIT Kanpur

• Worked on various projects related to image processing and computer vision.
• Current working on the project named: Towards Multimodal Assessment of Speaker Performance in Public Speaking. We propose a computational framework for quantifying speaker performance in the context of public speaking. We do not consider the content of the talk, rather analyze the speech and the visual content to capture the verbal and non-verbal behavior of the speaker. Inspired by the celebrated acceptance of deep learning frameworks over conventional methods, we are trying to model the speaker's behaviour using various deep neural architectures.
• In another of my recent works, I have developed a novel framework for segmenting crowd flow patterns from high density crowd videos. In this work, I have proposed an automated and completely unsupervised method to segregate crowd flows present in a video. The work has been accepted to appear in ICIP 2016.
• Apart from this, I have also developed a framework for calculating the Pixel-wise saliency values in an image based on color contrast and structure contrast of the image.

May 2015 – Jul 2015 · 2 mos · Greater Bengaluru Area

• Worked with the Imaging and Computer Vision group.
• Explored the implementation of Machine Learning Algorithms.
• Developed a generic library for 'Random Forest Regression' using C++. Also worked on the implementation of Discriminative Deep Metric Learning using Convolutional Neural Networks.

Dec 2014 – Dec 2014 · 0 mo · Greater Bengaluru Area

• Worked on the project named 'Cloud access for robot monitoring and support'. Explored solutions for providing both local and remote access to the robot controller using embedded wireless modules. Investigated use of Spark and Edison wireless modules and tested cloud access with both modules.

Jan 2014 – Jul 2014 · 6 mos · Greater Lucknow Area

• Worked on the software implementation of a Foot Mounted Inertial Navigation System, also named ‘OPENSHOE’ which is an open source software created by Signal Processing Lab, KTH Royal Institute of Technology, Sweden.
• Also created an Android Application which can communicate with the hardware of the OPENSHOE over Bluetooth socket and thus can collect the readings from the device and is further able to plot the walking paths of the user in real time.

May 2013 – Jul 2013 · 2 mos · Greater Delhi Area

• Designed their curriculum based on the special peer learning pedagogy introduced by Dr. Eric Mazur (Prof. with the Physics department at Harvard University).
• Created high quality conceptual tests for the aspirants of Engineering Entrance Exams such as JEE and AIEEE.

May 2013 – Jul 2013 · 2 mos

• The Learning Point is an initiative in open education, which aims to create a large scale and open repository of tutorials, quizzes and visualizations for students in high school and college.
• I created high quality tutorials relating College level Mathematics and Chemistry. These tutorials were included in the Learning Point’s repository of tutorials.