Data Science undergraduate building RAG systems, NLP pipelines, and scalable data applications.
I am a Data Science undergraduate specializing in Retrieval-Augmented Generation (RAG), NLP systems, and backend AI infrastructure. I build retrieval pipelines, evaluation frameworks, and low-latency AI applications focused on measurable retrieval quality and production reliability.
RAG system for real-time event queries deployed during a university event serving 400+ attendees. Reduced repeated-query latency from 4.2s to sub-20ms using multi-tier caching and semantic cache reuse. Led a 6-member development team across retrieval pipeline development, deployment, and testing.
Built a self-hosted cloud storage system with an integrated RAG pipeline for querying technical PDFs. Implemented a retrieval strategy using hybrid search, reranking, and context sufficiency checks to improve retrieval reliability and context grounding. Features asynchronous document ingestion, citation tracking, and context validation pipelines. Processed and indexed 1,000+ document chunks with hybrid retrieval and reranking pipelines to optimize retrieval quality.
Built a fault-tolerant news ingestion pipeline using dual-stage article extraction (Newspaper3k + custom DOM parsing) with NLP-based sentiment analysis and topic classification across live news sources.
Contributor to Keras — merged pull request reviewed by François Chollet addressing Apple Silicon MPS backend issues
