Anand Mudgerikar profile picture

Hi everyone! I'm Anand, I work on building large scale anomaly detection systems for enterprise Microsoft products like Azure Sentinel and Defender. I am part of the AMBITION AI team of the Security Research Division at Microsoft. Our team navigates a multiude of technical and research challenges, mainly using unsupervised Machine Learning for intrusion detection (obviously this is quite a hard problem in itself, if you think it's not so obvious, I would refer you to this nice article by Beetle and Sasha in the phrack magazine), while at the same time operating at a massive scale and within privacy compliance boundaries. Not quite sure how to succintly describe what I do here, Data/Research scientist? ML Engineer(official title)? Support/Data/Privacy Engineering?(not so fun but we do that plenty too!). Best way to put it is that I am a full stack ML Engineer with research expertise in Security and Machine Learning :)

Before Microsoft, I was a PhD student at Purdue University under the supervision of Dr. Elisa Bertino. My research interests included Information Security, Cryptography, Computer Networks and Machine Learning. I worked on developing 'safe' reinforcement learning frameworks for optimizing user QoS requirements and network functionalities in IoT environments (My Thesis). I was also part of the DAIS-ITA project which is a collaborative arrangement between U.S. and UK governments led by IBM for research in distributed analytics. We used transfer learning in conjugation with reinforcement learning to dynamically determine optimal policies in distributed network coalitions. I spent my summers of 2017 and 2018 at HPE Labs under the supervision of Dr. Puneet Sharma where I was involved in development of 'E-Spion: A system level intrusion detection system' and 'DITO: A honeypot service for IoT Devices'. Before pursuing my PhD, I also completed my Masters from the Center for Education and Research in Information Assurance and Security (CERIAS) with Dr. Elisa Bertino and Dr. Ioannis Papapanagiotou. My research and masters thesis were focused on development of hardware accelerated cryptographic libraries using GPUs for authentication in IoT environments.

Before coming to the states for my higher education, I completed my Bachelors in Information and Communication Technology from DA-IICT, India. For my Bachelor's thesis, I worked on improving the IPSec standard by incorporating secure multicast functionality using multi-party key computation under the guidance of Dr. Manik Lal Das. I was also working with TIFAC, Dept. of Science and Technology, Govt. of India with Dr. Prabhat Ranjan where I was responsible for conducting a research study to analyze the security threats in 3rd party applications and ensuring network security to prevent any unauthorized access.

For more details on these and other projects I am involved in, please check out the the projects tab. For a full list of publications, check out my google scholar page.