Joseph Rivas

Developed end-to-end NLP pipelines and production RAG workflows on AWS SageMaker, improving retrieval precision and scalable inference while implementing CI/CD and observability for LLM services.

Designed and deployed LLM-powered RAG and agentic systems on GCP to accelerate biomedical knowledge extraction and improved domain-specific Q&A accessibility for research workflows.

Built and maintained MLOps infrastructure on AWS, automated end-to-end ML pipelines, and deployed SLMs and containerized services to improve reliability, compliance, and time-to-production.

Developed and deployed NLP and CV models and scalable ETL workflows for aerospace applications, delivering production-grade AI solutions with MLOps, containerization, and data visualization.

Designed automated test methods and statistical validation workflows for surgical robotics, applying data-driven reliability analyses and supporting regulatory compliance and risk assessments.

Developed and automated test methods for medical devices, performed statistical analyses and root-cause investigations to ensure data quality and compliance with FDA and ISO standards.

Designed products using Pro/E and SolidWorks, performed FEA in ANSYS, and led prototyping and ECN coordination to deliver engineered solutions meeting ANSI/ISO and customer requirements.

Designed and executed experiments on magnetic thin films, performed SEM/EDX and profilometry, and produced reproducible data for actuator applications while gaining microfabrication and metrology experience.

Earned a Master of Science in Computer Science with advanced training in applied machine learning and research experience.

Completed a Bachelor of Science with foundational coursework supporting later engineering and research roles.