CLIENT WORK
LLMs IN HEALTH CARE
Developed clinical applications of Large Language Models for Vanderbilt University Medical Center. Project leveraged Django and multiple LLM libraries (Llama2, OpenAI, LlamaIndex, Replicate) to efficiently deliver AI-based patient summaries.
ML FOR MARKETING USE CASES
Served as Lead Data Scientist for Primer, collaborating with their CEO and Engineering team to develop and integrate multiple ML-based predictive models into their SaaS platform.
These models helped to:
• Model and identify high-conversion propensity accounts
• A/B test amount of lift from customers' use of Primer's platform
• Compare customers' campaign performance metrics (CTR, CPM, Reach) against data-driven benchmarks for Meta, Google Ads, and LinkedIn
• Predict target audience match-rates across Meta, Google Display/Search, and LinkedIn resulting in increased product reliability
PROJECT MANAGEMENT
My process in partnering with clients is both straightforward and effective:
-
Discussing and understanding your specific challenges and pain points is always the first step
-
Aligning on the scope of what a good solution or predictive model will look like and achieve
-
Building clarity around the availability and quality of the data within scope
-
Partner internally as needed to build solution while validating key steps and assumptions
-
Deliver solution using CI/CD conventions