Harsha S
Principal Software Engineering Manager at Microsoft
Redmond, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolHarsha S is a Principal Software Engineering Manager in Redmond with 17+ years of experience building secure, resilient cloud services and leading high-performing engineering teams at Microsoft. He combines hands-on systems and low-level engineering—contributing to open-source projects like DiskANN and EdgeML—with strategic leadership that drove major automation, AI adoption, and incident-reduction initiatives saving thousands of hours per build. Known for mentoring talent (14 promotions) and improving developer satisfaction and operational SLAs, he also holds a patent in access management and has a track record of shipping services with 99.9% availability. His background spans test automation, diagnostics, memory- and I/O-optimized ML code, and datacenter security configuration management, reflecting a rare blend of deep technical craft and measurable organizational impact.
9 years of coding experience
14 years of employment as a software developer
Bachelors, Electronics and Communications Engineering, Bachelors, Electronics and Communications Engineering at JNTUH College of Engineering Hyderabad
English
Github Skills (17)
algorithm10
algorithms10
c-language10
memory-management10
machine-learning-algorithms10
data-structure10
fileio10
file-processing10
file-handling10
data-structures10
file-access10
cprogramming-language10
edge-computing9
tensorflow8
pytorch8
Programming languages (5)
Github contributions (5)
microsoft/DiskANN
Jul 2020 - Jan 2023
Graph-structured Indices for Scalable, Fast, Fresh and Filtered Approximate Nearest Neighbor Search
Role in this project:
Back-end Developer
Back-end DeveloperContributions:4 releases, 263 reviews, 260 commits in 2 years 7 months
Contributions summary:Harsha contributed code to the `tsl/robin_hash.h` file, which is part of a graph-structured index. This likely involves implementing and maintaining core functionalities related to the graph's data structures and algorithms. The commits demonstrate involvement in low-level code, data structure, and algorithm. These changes indicate an emphasis on the index's internal logic.
graph-basedscalablenearest-neighborapproximate-nearest-neighbor-searchindices
microsoft/EdgeML
Aug 2017 - Dec 2020
This repository provides code for machine learning algorithms for edge devices developed at Microsoft Research India.
Role in this project:ML Engineer
ML EngineerContributions:2 releases, 34 reviews, 224 commits in 3 years 5 months
Contributions summary:Harsha contributed source code, specifically focusing on the `mmaped.cpp` file, which is part of a larger machine learning project for edge devices. This file deals with memory mapping functionality for reading and processing data. The code includes platform-specific implementations (Linux and Windows) using `mmap` and `CreateFileMapping` respectively, suggesting the user's involvement in low-level file I/O and optimization for different operating systems. The code demonstrates an understanding of data structures, file formats, and memory management techniques commonly employed in machine learning applications.
classifieredge-machine-learningsensortensorflowmicrosoft
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.