Prashant Chaturvedi

I’m an AI/ML Engineer with 5+ years of experience building production-grade ML and GenAI systems. I focus on end-to-end delivery—turning models into scalable solutions that are deployable and measurable in real time.

At G2 (Data Scientist II, Dec 2024–Present), I engineered a production-grade product categorization system that reduced support cases by 1,700/month and improved accuracy by 50%, exposed via real-time APIs. I also designed a scalable RAG pipeline using Milvus (embeddings, semantic search, and chunking), and built a review moderation system combining ML models, rule-based logic, and agent workflows to raise automation from 30% to 70%.

Previously at Tiger Analytics (Data Scientist, Mar 2022–Dec 2024), I developed a conversational AI chatbot for pharmaceutical experts to enable efficient knowledge retrieval. I built RAG systems using Qdrant with embeddings, hybrid retrieval, and prompt engineering, and created LLM-powered workflows with context-aware retrieval and structured outputs for enterprise use cases.

Earlier, as a Senior Analyst and Jr. Data Scientist, I led an end-to-end ML pipeline for cancer stage prediction using claims data, applying feature engineering and model selection (Logistic Regression, Random Forest, XGBoost) to reach an AUC of 85%. I also built an NLP/deep learning question generation system using POS tagging, embeddings, and fuzzy logic to improve quality and relevance, including faster educational content generation.