My name is Harini Kannan. I’m currently doing data science at Sophos AI. I have been in security data science for ~5 years. I was previously the Principal Data Scientist at Capsule8, a very exciting cybersecurity startup in NYC, that got acquired by Sophos (yay!). I have given talks at BlackHat (USA), Open Data Science Conference (East), Data Science Salon, PyData (Boston) and Data Connectors. I love writing about data science and applied machine learning in cybersecurity. My areas of research include:
- Using documented and undocumented performance counters to detect cache-side channel and other hardware based attacks like Spectre, Meltdown, Rowhammer and ROP
- Topic modeling for data discovery
- User and Entity Behavior Analysis
- ML for Cryptomining Detection
- Entity Resolution
- Graph Learning for anomaly detection
- Interpretable Machine Learning
- Autoencoders for unsupervised anomaly detection
