Feb 2019 – May 2022 · 3 yrs 3 mos

• I lead a team of research engineers at Gracenote, and we specialize in cross-domain Machine Learning (Computer Vision and Natural Language Processing). I help build algorithms that analyze Movies, TV Shows, Music, and Podcasts, by adding descriptive attributes (eg. genre, mood, subject), which enable better search and recommendation systems. I also help build systems that automatically generate appropriate imagery and descriptions for content, which enable better content discovery and user interfaces for our clients.
• My team is involved in the whole ML cycle - we shape products from ideation with the product team, build tools to collect datasets, build and validate the initial proof-of-concept, and work with engineering teams to deploy our algorithms.
• We publish and patent our work, I'm a named inventor on 7 issued patents, and most recently presented our work on Podcast Audio Summarization at the PodRecs workshop at RecSys-2020.

Dec 2017 – Feb 2019 · 1 yr 2 mos

• Working at the intersection of Data Science, Machine Learning, Signal Processing and Data Mining to develop novel algorithms for the next generation of Gracenote's products.

Jun 2014 – Nov 2017 · 3 yrs 5 mos

Jun 2013 – Aug 2013 · 2 mos · New York City Metropolitan Area

• Worked with the JUCE framework in C++ to develop Audio and MIDI processing functionality for the thinkplay application.

Aug 2012 – May 2014 · 1 yr 9 mos · Atlanta Metropolitan Area

• Masters in Music Technology

Jan 2012 – May 2012 · 4 mos · Bengaluru, Karnataka, India

• Worked on Compressed Sensing algorithms for Images and Optical Coherence Tomography.

Jun 2010 – Jul 2010 · 1 mo

• Projects on Wireless Sensor Networks and Spatial Audio Signal Processing.