New Jersey, United States
AI Engineer and Data Engineer developing production-grade machine learning systems, retrieval-augmented generation applications, and scalable data pipelines. Architected end-to-end AI solutions from data ingestion through model deployment and API integration.
Designed RAG systems using AWS Bedrock, ChromaDB, and LangChain, achieving 95%+ retrieval accuracy through optimized semantic chunking and vector embeddings. Built multi-phase pipelines covering ingestion, embedding generation, retrieval, and LLM-powered response generation with Claude models. Developed production chatbot processing 100+ documents with LangGraph conversation memory and PostgreSQL persistence, achieving <300ms query response times with citation tracking.
Developed NLP applications including sentiment classification systems using fine-tuned RoBERTa transformer models, achieving 89% accuracy on multi-class analysis. Implemented preprocessing and tokenization pipelines handling 1000+ reviews per batch, deployed as FastAPI services with <200ms inference latency.
Built ETL pipelines extracting data from 50+ counties via web scraping and US Census Bureau API, processing 10K+ records with comprehensive validation. Designed PostgreSQL schemas with 5+ staging tables, implemented retry logic with exponential backoff reducing API failures by 95%, and containerized applications with Docker for automated CI/CD deployments.
