PHANINDRA KETHAMUKKALA

About

LLM/RAG engineer with production deployments in highly regulated domains (Frost Bank and GE Healthcare). Built secure, explainable document-grounded Q&A systems using LoRA fine-tuning, strict RAG with confidence thresholds, and citation-based responses; also established evaluation/monitoring (golden QA sets, hallucination tracking, drift) and achieved ~40% latency reduction through retrieval/prompt tuning.

Experience

Education

Key Strengths

• Built and deployed production RAG system for internal document Q&A in banking with low hallucination tolerance
• Applied LoRA fine-tuning and domain-constrained prompting to handle domain-specific financial language
• Implemented strict grounding with confidence thresholds and citation-only answering to reduce hallucinations
• 4 years hands-on orchestration experience (LangChain + Python) integrating RAG workflows with FastAPI and cloud LLM endpoints
• Reliability-focused evaluation approach: unit tests, offline evals, drift/hallucination/task-success monitoring in production
• Iterative retrieval/prompt tuning using golden QA datasets; improved latency by ~40% while maintaining citation-backed accuracy
• Effective cross-functional collaboration with clinicians/radiology to align outputs with clinical terminology and safety guardrails