Sneh Vora

I’m an AI/ML Engineer with 4+ years of experience building intelligent solutions that improve document search, decision support, fraud detection, and risk analysis. I translate business problems into practical AI systems by taking models from data preparation through validation and deployment support.

Most recently, I built a RAG-based GenAI assistant for prior authorization and claims support using Python, LangChain, Azure OpenAI, and Azure AI Search. I developed document ingestion pipelines for 8,000+ pages, implemented embedding-based retrieval workflows (FAISS, metadata filters), and applied prompt engineering with guardrails, PHI masking, and RBAC controls aligned to HIPAA usage.

Earlier, at Accenture, I developed an AI-based fraud detection and risk scoring platform for a BFSI client. I engineered 40+ fraud-risk features, trained and validated an XGBoost classifier, used Isolation Forest for anomaly detection, handled imbalance strategies, and streamlined automated batch scoring with Python and SQL—improving model performance and reducing manual analysis effort.