Ravali Yerrapothu

About

Healthcare-focused ML/AI engineer who has built clinical note summarization and medical image annotation solutions using LLMs, RAG, and multimodal models. They combine experimentation across major model providers with practical production concerns like monitoring, drift detection, and latency/cost tradeoffs, and also earned 2nd place in a Google hackathon for a medical AI assistant.

Key Strengths

• End-to-end ownership of production healthcare LLM summarization systems
• Strong evaluation and monitoring mindset for GenAI systems
• Applied RAG and retrieval-based approaches to clinical workflows
• Healthcare AI experience spanning clinical notes, prior authorization, and medical imaging
• Research-to-production translation in time series seasonality modeling
• Pragmatic latency/cost optimization using persisted simulation outputs
• Scalable backend implementation across AWS and Java Spring/Hibernate stacks
• Builds reusable Python utilities that improve engineering efficiency
• Hackathon recognition: won 2nd place in a Google competition for MedAnnotator
• Built and deployed an LLM-powered clinical note summarization system for patient/provider/payer conversations
• Designed hybrid retrieval (keyword + semantic) to resolve discrepancies and improve medical-code-precise retrieval
• Grounded outputs against a medical code database to reduce terminology/coding mistakes
• Hands-on orchestration with LangChain (tokenization, templates, RAG/search criteria) and LangGraph (agent-based summarization/Q&A)
• Pragmatic evaluation approach using human review and repeated-input consistency testing
• Able to align non-technical stakeholder expectations with data/feasibility constraints (Virtusa/Comcast sales prediction use case)

Experience

Education

Awards

• Runner up in Agentic AI hackathon - Google Cloud & ODSC
• Orion Eminence and Excellence Award - IBM (2014, 2015)
• Deep Dive with skills - IBM (2015)
• Employer’s Choice Award - IBM (2014)

Skills

Languages

Certifications