Greater Seattle Area
I'm Madhav ( mɑːd̪ʰəv ) a Software Engineer who is passionate about integrating machine learning into real-world engineering applications. I am currently building cloud and ML infrastructure powering programmatic advertising at Moloco. I have published research papers, presented tutorials, and served as conference reviewer at leading computing conferences such as ACL, InterSpeech, IEEE Transactions on Audio Speech and Language Processing, MLSys, and SciPy. Most recently, I wrapped up my Masters Thesis from the University of Washington focusing on GPU system optimizations to speedup large-scale LLM deployments. Check out my work on my Github: https://github.com/madhavmk Email me at: [email protected]
Building Cloud and Machine Learning Infrastructure for Programmatic Advertising. Working on the Demand Side Platform (Measurements Team).
* Developed ”Seafloor Geodesy”, an open source scientific tool used in measuring seafloor crust deformations. My role includes processing large scale Sonar and Satellite (GNSS) data using Numpy and Pandas, speeding up scientific algorithms using Numba just-in-time compiler for Python, and creating intuitive visualizations for earth scientists. * Presented ”Generative AI Copilot for Scientific Software – a Retrieval Augmented Generation Based Approach” Tutorial at the SciPy 2024 conference. Optimized an OLMo model for on-device edge CPU inference using quantization and llama.cpp.
* Working in the core backend team of the zero trust cybersecurity product "Enterprise Application Access". My responsibilities include end-to-end delivery of features enhancing network security and increasing scalability. * Implemented Django middleware to reduce master database load by rerouting SQL queries to alternate eventually consistent read-only replica databases. Resulted in a 40% reduction in the primary database’s CPU load. * Designed agent inline security upgrade pipeline to keep network-isolated systems up to date with the latest Debian package patches. My pipeline assisted with the Ubuntu 18.04 to 22.04 OS upgrade on 5,000+ running production agents. * Led development of configurable DNS feature enabling customers to set which public and internal DNS upstream servers their agents utilize. The dnsmasq based design worked on heterogeneous agent platforms including Virtualbox, Azure etc.