Blake Martin

I am a Senior Python AI/ML Engineer with over a decade building production-grade machine learning systems and six-plus years focused on Generative AI, LLMs, and RAG. I turn large-scale structured and unstructured data into reliable, measurable business value across healthcare, fintech, and edtech.

At Tempus I led development of LLM-powered clinical decision support systems, building Hugging Face transformer pipelines and RAG solutions that improved clinical concept accuracy and reduced hallucinations. I designed FastAPI microservices for sub-second inference and automated LLMOps pipelines to compress deployment cycles from weeks to hours.

At Plaid I developed transaction classification and risk models, large-scale PySpark feature pipelines, and low-latency inference services, improving accuracy and reducing latency and compute costs. Earlier roles included recommendation and NLP systems that boosted engagement and completion rates through well-engineered feature pipelines and retraining workflows.

I prioritize scalable, observable MLOps/LLMOps, efficient cloud-native deployments (AWS, GCP, Azure), and cross-functional collaboration with clinicians, product, and platform teams to align models with real-world workflows and regulatory needs.