Sahil Garg is a Principal AI Researcher leading AI research for offensive and defensive security, spanning LLM post-training, reinforcement learning and continual learning, with applications in penetration testing, vulnerability discovery, threat detection and remediation. With over 1,500 citations, he has published at ICML, AAAI, EMNLP, UAI and IJCAI and in Nature Partner Journals (npj). He has applied machine learning to high-stakes problems in cancer biology (PhD, USC), computational psychiatry (postdoc, Mount Sinai) and finance (Vice President, Machine Learning Research, Morgan Stanley), where he co-authored the time-series foundation model Lag-Llama. His focus is robust, data-efficient deep learning.