Dhivya Senthil Murugan
Engineer Ii @Medtronic
Signup · Get unlimited contacts
WORK HISTORY
Engineer Ii @Medtronic
Designed and implemented a self-service analytics pipeline using OpenAI API and Snowflake, allowing non-technical users to query data in natural language- Led analyses to support FDA submissions, generating requirements from cross-functional stakeholders and iterating at a fast pace to deliver a meaningful final analysis- Predicted customer attrition by analyzing patient device data, visualizing patient journeys, and supporting strategic marketing decisions- Collaborated cross-functionally with systems engineering, customer-centered design, clinical safety, regulatory, and marketing teams to align technical solutions with business goals- Mentored and led a team of interns, managing technical projects, delegating tasks, and guiding their professional growth.
EDUCATION
Georgia Institute of Technology
Master's degree, Computer Science (Machine Learning)
University of Southern California
Bachelor's degree, Bioengineering and Biomedical Engineering
ABOUT DHIVYA SENTHIL MURUGAN
I build machine learning systems that improve patient outcomes — and I\'ve been chasing that intersection my whole career.I\'m a computer scientist with a heart that beats for medtech. I started at USC as a biomedical engineering major, but I switched into computer science and computational biology; I believed the next frontier in healthcare would be driven by the algorithms and data behind the devices. That conviction led me to a Master\'s in Machine Learning at Georgia Tech (earned while working full-time) and nearly five years at Medtronic, where I\'ve been the ML engineer embedded across clinical, regulatory, and product teams.At Medtronic, I\'ve built an NLP-powered analytics pipeline that lets non-technical teams query clinical datasets in natural language, predictive models that catch at-risk patients before they disengage, and customer segmentation that directly shaped product design. I\'ve shipped internal tools from zero to production in 3–4 days, each saving teams 10+ hours per week. I\'ve led analyses for FDA submissions and mentored 10+ interns along the way.Outside of work, I build the projects I wish I could build everyday. TrialTwinLab is a digital twin pipeline for oncology clinical trials. I fine-tuned BioBERT for clinical named entity recognition — extracting diseases and drugs from prior authorization notes — and stress-tested the domain transfer gap between PubMed and real clinical text. I built a dental X-ray classifier using transfer learning on a tiny dataset and deployed it as a serverless inference endpoint on AWS Lambda. Each one taught me something different about what it takes to move ML from notebook to production in healthcare.My next chapter will be as an ML Engineer, AI Engineer, or Clinical Data Scientist where machine learning is the core product, not a support function. I want to be on a team where the quality of the model directly determines the quality of what reaches the patient.Check me out at https://dhivya.space/, and looking forward to connecting!
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.