Applied Scientist at Amazon Contextual Ads. Working on building LLMs to improve Contextual Ad Targeting systems.

Applied Scientist at Amazon Search Science and AI. Working on building large-scale transformer-based models for Search and Ads applications.

Applied Scientist at Amazon Search Science and AI. Working on building large-scale transformer-based models for Search and Ads applications.

• Worked as an Applied Scientist Intern at Amazon Search.• Developed deep learning models to link context-of-use entities with products to improve the search experience of products.• Proposed BERT-based and co-teaching approaches to improve precision and coverage of links compared to lexical matching while also overcoming noisy labels in training data.• Work accepted at the Multi-faceted Understanding of Products and Shopper Intent Workshop at AMLC 2020.

• Worked in Visual Learning and Intelligence (VIGIL) Lab at IITH under Dr. Vineeth Balasubramanian. • Developed a novel Multi-Space model for Zero-Shot Object Detection (ZSD).• Leveraged both semantic and visual spaces and introduced a cross-modal consistency loss to alleviatethe hubness problem in ZSD.• Model outperformed the state-of-the-art in ZSD on Pascal VOC by 14% in terms of mAP.• This work has been accepted at WACV 2020.

• Developed a plug and play service for effective management, monitoring and usage of Test clusters for the Azure Networking Team.• Facilitated easy check-in (locking) and check-out (unlocking) of clusters and devised health checks for the maintenance of clusters.• Built a UI dashboard alongside the service for interacting and reporting.

• Worked at Institute of Systems Studies and Analysis (ISSA), DRDO in the domain of Data Mining and Associative rule learning.• Worked on Frequent Itemset Mining algorithms for large certain and uncertain datasets.• Extended the Eclat algorithm and developed FT-Eclat algorithm for Fault-Tolerant Frequent Itemset Mining.