Jason Krause

Led development and deployment of production LLM and ML pipelines, shipping RAG-based features that improved workflow efficiency by 32% while implementing hallucination detection and evaluation frameworks to meet compliance standards.

Advised and prototyped production-ready GenAI systems, designing RAG and agent-style orchestration workflows and building benchmarking harnesses to evaluate latency, throughput, and quality trade-offs.

Built scalable ML platforms and distributed training pipelines for 100M+ records, developed cloud-native ML services and CI/CD workflows, and optimized inference stacks for latency and portability.

Applied ML research in NLP and computer vision, developing classifiers, OCR pipelines, and GPU-based PyTorch training to support reproducible experimentation for M.S. research.

Developed and productionized real-time fraud detection ML systems, improving precision by 22% and delivering low-latency inference services (<100ms) with automated monitoring for drift.

Prototyped NLP classifiers for medical document classification, built scikit-learn models, and contributed annotation scripts and evaluation for biomedical text tasks.

Completed a Master of Science in Computer Science while conducting research and engineering work in NLP and computer vision.

Completed a Bachelor of Science in Statistics with coursework supporting data analysis and applied statistical modeling.