Senior Data Scientist at CVS Health. I build and study agentic AI systems in production and in research. At CVS Health (Fortune 5), I've architected multi-agent LLM systems for medical policy retrieval and ML models for claims overpayment detection. As a researcher, I'm running empirical studies on how agentic AI fails, with results that contradict what most practitioners assume about AI self-correction. MS, Columbia University and BTech, IIT Madras. Available to comment on: agentic AI, LLMs in production, healthcare AI, ML in high-stakes environments.