Yusmitha Lekha Prathi

I am an AI and ML engineer with four years of experience building production ML systems for healthcare and finance. I design and deploy LLM-powered workflows and scalable ETL pipelines to deliver measurable operational improvements.

At Capital RX I built an LLM multi-agent workflow for prior-authorization and claims analysis, deployed on Kubernetes, improving processing efficiency and reducing manual interventions. I implemented RAG pipelines, vision-language extraction on Vertex AI, and observability with LangSmith to ensure traceable, safe model behavior.

Previously I developed batch ETL and SQL transformation pipelines for banking data at NatWest via HCL, and created interactive Power BI and Tableau dashboards at AGS Health that reduced turnaround times and improved claim accuracy. I focus on actionable KPI insights, data quality, and HIPAA-compliant analytics.

I aim to leverage LLMs, ML engineering, and cloud-native deployments to create secure, compliant AI systems that produce measurable business value for clinicians and financial stakeholders.