Sep 2023 – Present · 2 yrs 7 mos · San Diego, California, United States

Mar 2022 – Sep 2023 · 1 yr 6 mos

• Azure Linux Systems Group

May 2021 – Mar 2022 · 10 mos

• Azure Linux Systems Group

Sep 2019 – Apr 2021 · 1 yr 7 mos

• Engineering Systems Team
• Worked on support and upgrade of engineering infrastructure for virtualized artifact packaging in Windows OS build.
• Worked on the design and development of PowerToys Run, a tool to quickly launch windows apps without sacrificing performance.

Jun 2019 – Aug 2019 · 2 mos · Bengaluru Area, India

• AI Labs
• Worked on Bayesian online changepoint detection (BOCPD) in a multivariate time series modeled using Gaussian Process
• Improved complexity of algorithm from O(n^2) to O(n) for a class of Covariance function making it feasible for real-time detection.

Jan 2019 – Apr 2019 · 3 mos · Mumbai Area, India

• Guide: Prof. Arjun Jain, CSE Department
• Formulated method to capture aleatoric uncertainty in occluded human pose estimation.
• Established quantitative results on the propagation of uncertainty between connected joints in the presence of occlusion.
• Significantly improved 3D triangulation of multi-view images by weighing joints with their uncertainty.
• Publication accepted at CVPR '19 workshop on uncertainty in Deep Visual Learning.

May 2018 – Apr 2019 · 11 mos · Mumbai Area, India

• Courses: Operating System (CS 347), Database and Information Systems (CS 317)
• Contributed to designing and evaluating labs, assignments and preparing class tests

May 2017 – Apr 2018 · 11 mos · Mumbai Area, India

• During my tenure, I revamped the SMP website with the aim to inform prospective student and freshers about culture and academics at IIT Bombay. The website resulted in 80k+ student views across the country.

Jun 2018 – Aug 2018 · 2 mos · India

• Mentored a student in developing distributed graph algorithms

May 2018 – Jul 2018 · 2 mos · Hyderabad Area, India

• Microsoft India Development Centre | Microsoft IPE
• Designed and implemented a solution to find the origin of error in the job pipeline, starting with erroneous files.
• Created a pre-processor to parse the job graph and create daily map of input and output files of each job.
• Created a query engine that takes in output files and lists corresponding jobs and input files.
• Used D3.js and ASP.NET to create a web-app that facilitates tracking pipeline interactively.

May 2017 – Aug 2017 · 3 mos

• Contributed distributed algorithms for centrality measures and shortest path
• Implemented Graph500 benchmarks to rank distributed systems on massive graph operations.
• Presented poster in JuliaCon 2017 at UC Berkeley on performance benchmarks of parallel Betweenness centrality algorithm run on Scale-free and Erdos-Renyi graph

Dec 2016 – Dec 2016 · 0 mo · Bengaluru Area, India

• Used parallel constructs to analyze trends in New York taxi dataset for a period of 29 months.
• Cleaned the dataset from erroneous trips and plotted city heat-map for pickup and dropoff.
• Used clustering algorithms to analyze shift in prime pickup locations