Hello! I am Mahadevan. I am a software engineer by day (and by night). Coding is my passion.

I currently work as a Software Engineer at Google in Irvine, California, as part of the Google Analytics team. My work focuses on designing privacy-safe measurement techniques, building end-to-end machine learning systems using TFX and TensorFlow to model user behavior, and conducting large-scale data analysis with Apache Beam pipelines. I also lead our cross-channel budgeting backend services and develop predictive models for Ad Spend and Revenue using Media Mix Modeling (MMM).

Previously, I was a Senior Software Engineer at Microsoft in Bellevue, Washington, on the Speech Runtime team. I led the integration of ONNXRuntime for offline and online speech recognition inferencing scenarios and designed features to measure user-perceived latency. I also served as the module lead for the Keyword Spotting Runtime team, where I implemented cross-platform support for various DNN/CNN models and built the runtime for Microsoft’s RNNT-based Custom Keyword Spotting engine (for which I co-authored a patent for on-device custom wake word detection).

Before Microsoft, I worked for the Location Technologies team at Qualcomm, where I developed algorithms to search and track GNSS satellites (for which I hold a patent).

I hold a Master’s in Electrical Engineering from Pennsylvania State University, where my thesis focused on neural network models for predicting a patient’s conversion to Alzheimer’s disease based on MRI data (full thesis here).

Courses & Certifications