Feb 2021 – Nov 2022 · 1 yr 9 mos · Bengaluru, Karnataka, India

• Worked as a member of the SaaS team which specializes in Microsoft Office365 component's Data Management (Backup, Recovery, and other related miscellaneous workflows). Skills- Distributed Computing, Asynchronous programming, Low level programming, C++, Angular, Javascript, Golang

Jun 2020 – Aug 2020 · 2 mos · Kaiserslautern, Rhineland-Palatinate, Germany

• Optimizing Covid-19 group testing process via contact tracing
• Developed group testing algorithms to minimize the number of tests required to test a given number of people.
• Employed location and proximity-based contact tracing data available via satellites and tech services to compute the accurate prior probability of infection for every individual based on his mobility and exposure to infection.
• Achieved reduction in the number of tests by up to 87%for data on Kaiserslautern, Germany

Mar 2020 – Jun 2020 · 3 mos · California, United States

• Object Detection in Augmented Reality
• Implemented several state-of-the-art object detection algorithms namely SSD, BlazeNet, RetinaNet, CornerNet, CenterNet, Transformers, etc. optimized to work in AR setup to estimate apparel sizes based on short videos.
• Deployed the models on mobile devices using MobileNet-V2 as feature extractor underneath these single-shot detectors

Nov 2019 – Mar 2020 · 4 mos · Singapore

• Anomaly detection using Causal Inference
• Developed causal graphical models to solve the problem of anomaly detection in multivariate time series data.
• Identified Granger Causal Relationships among various variables and used them to explain and infer hidden anomalies.
• Improved existing performance onSWaTandWADIdatasets in an interpretable and scalable manner.

Aug 2019 – Nov 2020 · 1 yr 3 mos · Kanpur, Uttar Pradesh, India

• 2019-20-I Teaching Assistant with Prof. Sumit Ganguly, ESO207A(Data Structures and Algorithms)
• 2019-20-II Tutor with Prof. Nisheeth Srivastava, ESC101A(Introduction to Programming)
• 2020-21-I Teaching Assistant with Prof. Anil Seth, ESO207A(Data Structures and Algorithms)
• 2020-21-II Teaching Assistant with Prof. Raghunath Tewari, ESO207A(Data Structures and Algorithms)
• Designed the course structure, exam questions, tutorials and planned other logistical/technical events for courses involving 500+ students.

May 2019 – Jul 2019 · 2 mos · Gurgaon, India

• Predictive Reviewer Profiling
• Developed a machine learning model to predict the credibility of new users on Gartner's Peer Insights platform.
• Improved efficiency of the review moderation process by∼47%leading to reduction in costs.
• Conceptualized badging of the reviewers based on their credentials and the quality of reviews written by them.
• Identified strategic clusters of users for targeted campaigning using Jensen - Shannon divergence.

May 2018 – Aug 2018 · 3 mos · IIT Kanpur, India

• Joint Audio-Visual Generation and Discrimination, Supervisor: Prof.Vinay P. Namboodiri, IIT Kanpur
• Developed a multimodal dataset-CAMP-MNIST of size 0.5 million for combined generation of audio, video & text.
• Improved appearance of generated images by incorporating a GAN learnt latent space combining GANs and VAEs.
• Conceptualized and designed an architecture for conditional cross-modal generation & alignment in PyTorch.
• Analyzed variants of our model by varying convolution dimensions, softmax temperature, kernel size, architecture
• depth & hyper-parameters.

Feb 2018 – Jun 2018 · 4 mos · Bangalore, India

• Predicting Stock Prices to Develop Trading Strategies for different stock market indices
• Developed predictive models for stock prices in Python using the fundamentals of quantitative finance research.
• Designed, back-tested and optimized a data-driven quantitative trading strategy on real-world data in python.
• Developed an intra-day mean reversion strategy to give >30% return on capital(RoC) using Hurst and ARIMA.
• Built a consistent and unified framework to forecast sensor parameters and measure uncertainty in them

Dec 2017 – Apr 2018 · 4 mos · Kanpur Area, India

• Applied Machine Learning models to detect equipment malfunctions and anomalies/outliers in sensor data
• Processed data pipeline to fill communication gaps in transmission of data for water quality measurement