Harish Reddy is a software engineer with 12 years of experience focused on video codecs and systems engineering, currently building Windows-related solutions at Microsoft after a strong tenure in embedded and codec development. He has deep hands-on expertise in VP9 and other video codecs, contributing performance optimizations, neon assembly integration, and row‑based multi-threading work that improved decoder stability and throughput. At OSS-Fuzz he integrated fuzzing targets and seed corpora for multiple video codecs, demonstrating practical security-minded build and release automation for widely used open-source projects. Based in Bengaluru, he blends low-level performance tuning with CI/build engineering, and brings a track record of shipping robust multimedia code from Ittiam and Samsung R&D to global platforms. An early clue to his practical mindset: his GitHub activity shows he prefers improving reliability and safety in foundational libraries rather than flashy demos.
12 years of coding experience
Bachelor of Technology (BTech), Computer Science, First class with honours, Bachelor of Technology (BTech), Computer Science, First class with honours at PES University
+2, Computer Science, 96%, +2, Computer Science, 96% at christ junior college
OSS-Fuzz - continuous fuzzing for open source software.
Role in this project:
Automation Engineer / Build & Release Engineer
Contributions:11 commits, 25 PRs, 27 comments in 11 months
Contributions summary:Harish's contributions primarily revolve around building and integrating fuzzing targets within the OSS-Fuzz environment. They focused on adding fuzzers for various video codecs like libvpx, libaom, libavc, libhevc, libmpeg2, libxaac, and libultrahdr, often involving modifying build scripts to include necessary dependencies and configuration options. Their work also involved integrating seed corpora and dictionaries for the fuzz targets, demonstrating a practical understanding of fuzzing workflows.
Contributions summary:Harish contributed to the VP9 video codec library, specifically implementing and modifying components related to decoding and motion compensation. Their work involved adding a flag to enable row-based multi-threading for VP9 decoding and making corresponding modifications to the decoder interface and related structures. The user also made changes to integrate new Neon assembly routines to optimize motion compensation operations, enhancing the performance of the codec. Additionally, the user addressed issues related to incorrect handling of skip in row-mt mode, thereby improving the overall decoding stability and efficiency.
pull-requests
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.