At Blackboard, I built LangGraph agents with Anthropic Claude to identify at-risk students and automate personalized outreach, reducing instructor follow-up by 12 hours per week. I also implemented RAG pipelines that cut course discovery time from two minutes to under 30 seconds.
At Medtronic, I built a personalized insulin-dosing platform with Java, Spring Boot, and React that supported $2.2B in annual revenue. I improved patient-message classification accuracy to 91% by fine-tuning healthcare NLP models, and created an ADA-compliant component library that reduced accessibility QA defects by 95%.
Earlier, at Boston Consulting Group, I supported Axle’s $10 million Series A fundraising by re-architecting a Django monolith into microservices. At Spotify, I improved backend services supporting 75 million users and reduced inter-service latency by 80%.

