Vamsi Manchala
ML Compiler Engineer at Google
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolVamsi Manchala is an ML Compiler Engineer at Google with a strong foundation in Modern C++ and three-plus years of hands-on experience building performance-critical software for autonomous vehicles, robotics, and real-time embedded systems. He brings deep expertise in ROS2, Apex.OS, CyberRT and DDS implementations, applying advanced C++ patterns—memory management, concurrency, and async design—to hard real-time and safety-critical problems. His background includes impactful work at Ford and MathWorks and academic projects in SLAM, localization, and deep learning, giving him a practical bridge between research and production. Vamsi is an active contributor to TensorFlow/TFLite projects where he improved converter lowering, quantization handling, and TFLite Micro logging/error reporting—skills that reflect his focus on making ML models run efficiently on constrained embedded targets. Based in Sunnyvale, he combines systems-level rigor with curiosity about emerging tech like blockchain, continually exploring new ways to solve complex engineering challenges.
3 years of coding experience
11 years of employment as a software developer
Master of Science (MS) Engineering, Master of Science (MS) Engineering at Arizona State UniversityRobotics Software Engineer Nanodegree, Robotics Software Engineer Nanodegree at UdacityBachelor of Technology (BTech) Electronics and Communication Engineering, Bachelor of Technology (BTech) Electronics and Communication Engineering at Visvesvaraya National Institute of TechnologyDeep Learning, Deep Learning at Harvard Extension School
English, Hindi, Telugu
Github Skills (19)
machine-learning10
mlr10
tensorflow-lite10
tensorflow10
sys10
tflite10
embedded10
c-language9
stable9
microcontroller9
cprogramming-language9
quantization9
python8
deep-learning8
neural-network7
Programming languages (3)
Github contributions (5)
tensorflow/tflite-micro
Sep 2022 - Dec 2022
Infrastructure to enable deployment of ML models to low-power resource-constrained embedded targets (including microcontrollers and digital signal processors).
Role in this project:
ML Engineer
ML EngineerContributions:51 reviews, 18 commits, 21 PRs in 3 months
Contributions summary:Vamsi primarily contributed to the refactoring and modification of the TFLite Micro framework, specifically concerning the logging mechanism and error reporting. They introduced changes to move the `MicroPrintf` logging utility function to `micro_log.h` and `.cc` for better organization. The user also updated examples to remove the usage of `ErrorReporter` in favor of `MicroPrintf`. Furthermore, the user made changes to the framework classes to eliminate the usage of `MicroErrorReporter` and `ErrorReporter` classes.
signalml-modelslow-powerprocessorsdeployment
tensorflow/tensorflow
Aug 2022 - Apr 2025
An Open Source Machine Learning Framework for Everyone
Role in this project:ML Engineer
ML EngineerContributions:11 reviews, 1 comment in 2 years 8 months
Contributions summary:Vamsi's commits primarily involve the development and optimization of machine learning models within the TensorFlow framework. Their contributions include creating and testing patterns for lowering composite operations, like those related to JAX and PyTorch, to TFLite operators. The user also worked on improving the conversion process, including the handling of constant folding, and supported the use of per-axis quantization for operations like transpose and reshape. Furthermore, the user added support for more operations to be handled by the TFLite converter, specifically focusing on operations within the stablehlo dialect and also contributed to the improvement of FlatBuffer export performance.
pythondata-sciencedeep-learningmlmachine-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.