Mar 2021 – Jan 2022 · 10 mos · San Jose, California, United States

• At Nutanix, as part of the File Analytics team, I designed and planned AWS infrastructure for distributed processing of events in file servers based on Spark and Kubernetes.
• I also helped lay out the foundation for ML-based service offerings in public cloud and drove the development of models for anomaly detection.

May 2020 – Aug 2020 · 3 mos · San Jose, California, United States

• My project's objective was to implement robust Machine Learning models for anomaly detection in operations performed on large file servers.
• I built a variant of the Holt-Winters Model for temporal based anomaly detection on per hour operation counts with adaptive thresholding
• I also implemented and used Collaborative Filtering for detection by learning user behaviour using peer comparison for file access patterns

May 2018 – Jul 2018 · 2 mos · Bengaluru, Karnataka, India

• As part of my project at Goldman during my 2018 summer internship, I built a light calculator for customer margin allocation for proper internalization based on a waterfall logic, taking various parameters into consideration and a greedy strategy to allocate margin per stock, improving the run time of the allocation logic by tenfolds.
• Other than this, I introduced changes in existing models for asset-liability gap management identifying bugs in pre-existing code, and augmenting features for proper AL Gap Management for deposits and clearinghouse initial margin.

May 2017 – Jul 2017 · 2 mos · Bengaluru, Karnataka, India

• I worked as an intern in Inmobi for the summer of 2018 where I worked on extracting image features from Ad creative images using OpenCV (in python) and Google Cloud Vision API to explain ad Click Through Rate (CTR). I analysed Pearson Correlation, performed regression tests, and used variable selection in Weka to detect explainable features out of which a set of actionable features were extracted. Moreover, I created a python server to handle feature extraction and prediction for building suggestions for possible Ad enhancements based on CTR.

Jan 2016 – Jul 2016 · 6 mos · Greater Delhi Area

• Acted as the lead architect of the core content engine and front web interface. Other than that, I designed and developed a dashboard for the content creators to add and edit questions using PHP and MySQL.

Dec 2015 – Dec 2015 · 0 mo · Bangalore, India

• Developed a backend Curator Tool which scraped, rendered and compiled information from the web using Beautiful Soup in Python during my winter internship at Inmobi.