At Cognizant Technology Solutions, I tested AI-generated outputs and LLM workflows for TriZetto, including RAG pipelines and agent-based workflows. I also evaluated model responses for relevance, accuracy, fairness, and bias.
For SynAI implementations, I engineered and tested prompt variants, chaining, and few-shot and zero-shot approaches. I documented evaluation findings and performance outcomes to support model refinement and go/no-go decisions.
Across healthcare projects, I tested Facets systems, claims, enrollment, and 834 and 837 EDI transactions. My work included testing membership and claims migration, benefits, and production deployment releases.
Earlier, I tested Guidewire Claim Center workflows for QBE and loan processing for Westpac. I also hold certifications in Introduction to Prompt Engineering and Prompt Engineering with ChatGPT.
