Pranav Sharma
Senior Staff Software Engineer at Meta
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolPranav Sharma is a Senior Staff Software Engineer in San Francisco with 25+ years of hands-on experience designing and building high-performance, large-scale C++ runtime and serving systems across advertising, search, e‑commerce, and AI. He helped found and lead ONNX Runtime development at Microsoft and drives its Triton backend work, contributing core graph processing, operator implementations, and type/shape inference improvements to widely used open-source ML runtimes. Previously he architected mission-critical ad-serving and real-time indexing systems at Yahoo and led conversational bot runtime platforms, demonstrating deep expertise in low-latency distributed systems and production ML serving. Comfortable both as an individual contributor and technical lead, he blends security-hardened engineering from early roles with modern AI-serving optimizations. An uncommon combination of core systems performance tuning and ML interoperability makes him especially effective at taking model inference from research into robust, scalable production.
10 years of coding experience
14 years of employment as a software developer
M.S. (Master of Science) Computer Science, M.S. (Master of Science) Computer Science at Syracuse UniversityB.E. (Bachelor of Engineering) Electronics, B.E. (Bachelor of Engineering) Electronics at University of MumbaiHacking by Numbers (Bootcamp edition), Hacking by Numbers (Bootcamp edition) at Sensepost (http://www.sensepost.com)SEI Software Architecture Professional Certificate Software Architecture, SEI Software Architecture Professional Certificate Software Architecture at Carnegie Mellon UniversityMachine Learning (Applied - CS229A), Machine Learning (Applied - CS229A) at Stanford University (Coursera)
Stackoverflow
1,095reputation
110kreached
8answers
5questions
Github Skills (24)
c-language10
operation10
tensorrt10
machine-learning10
inference10
onnx10
deep-learning10
tensorflow10
tensor10
cprogramming-language10
shapes10
data-structure9
algorithm9
pytorch9
algorithms9
Programming languages (11)
Github contributions (5)
microsoft/onnxruntime
Nov 2018 - Jan 2023
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1689 reviews, 325 commits, 451 PRs in 4 years 2 months
Contributions summary:Pranav primarily worked on improving the ONNX Runtime's core logic, focusing on the graph processing and operator implementations. Their contributions included modifications to core files related to graph creation, node processing, and type inference, as well as new implementations of operators such as OneHot and Det. The user also participated in refactoring and optimizing existing code for efficiency and correctness. Moreover, the user added support for the String type and added support for handling more than one input to the Tensor Sequence.
runtimetrainingtensorflowai-frameworkaccelerator
onnx/onnx
Apr 2018 - Mar 2020
Open standard for machine learning interoperability
Role in this project:ML Engineer
ML EngineerContributions:1 review, 9 commits, 7 PRs in 1 year 11 months
Contributions summary:Pranav contributed significantly to the ONNX project by enhancing the framework's capabilities for shape and type inference, specifically focusing on operators related to instance normalization, LpNormalization, and IsNaN. Their work involved modifying operator schemas, implementing shape/type inference functions, and adding relevant test cases within the shape inference testing framework. Furthermore, the user made adjustments to the code to prevent warnings related to unused variables and contributed to clarifying the specification of convolution and convolution transpose.
pytorchmxnetdeep-learninginteroperabilitymachine-learning
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.