Working on 3D object detection. Duties include reviewing relevant literature, design and implementation of the state of the art algorithms and deep learning network architectures for 3D object detection.

Working on 2D detection using RGB/RGB-D. Duties include reviewing relevant literature, design and implementation of the relevant algorithms and deep learning network architectures.

Deep learning models are extensively used in various safety-critical applications. Hence these models along with being accurate need to be highly reliable. One way of achieving this is by quantifying uncertainty. The uncertainty quantification methods have been extensively studied for deep learning models applied to images but have been less explored for 3D modalities such as point clouds often used for autonomous systems. As part of this work, we adapt and evaluate uncertainty estimation approaches on 3D semantic segmentation models. We perform a state of the art analysis of 3D semantic segmentation models working on point clouds and shortlist two methods: DarkNet21Seg and RandLA-Net based on their performance in the SemanticKITTI dataset. Three uncertainty quantification methods namely Deep ensemble, MC-Dropout and MC-DropConnect for DarkNet21Seg and Deep ensemble, MC-Dropout and Test time augmentation for RandLA-Net are evaluated to study the impact of various parameters such as the number of models in ensembles, dropout/dropconnect value, and the number of forward passes on various metrics which take into account the performance, uncertainty estimates, and reliability of the model.

Planned and organized a three week long bridging for new students joining the MAS course in Hochschule Bonn Rhein Sieg to get them familiarized with the concepts required for the course. During the course, I have also taught different core modules such as Python, Git, Java and Data structures.

Below are a few of the details on the work/responsibilities I took up during projects I worked on as a Senior Systems Engineer.1. Worked extensively on analyzing and building two Granular Rest Services which formed the backbone of the entire project.2. Rectified critical issues related to Granular services, which were caught during testing, and guided a team of 3 people to resolve these issues.3. Coordinated with the Bangalore and Illinois, USA teams under the Agile SDLC.4. Took up various KT sessions for new joiners to get acquainted with the project and technical approaches.

Worked on enhancing the Web systems that provide a diverse range of business functions such as policy administration and tools to support the sales and servicing of policies under Waterfall SDLC. Developed a thorough understanding of major design patterns while working on these projects.

Underwent a five-month long training in Infosys Ltd. The training covered topics such as Operating Systems, Algorithms, OOPS concepts and JAVA, Data Structure, DBMS, and Software Engineering. I was ranked as a High Performer during the training