Sambodh Gupta

I’m a Computer Science and Engineering B.Tech student building real-world data products—especially pipelines that automate work, improve data quality, and make results understandable to people.

As a Data Analyst Intern, I automated an end-to-end social media posting workflow using Meta Graph API and Google Apps Script, removing ~8 hours/week of manual scheduling. I cleaned and standardized 20,873 records across a 409-field schema, fixing 12 categories of data quality issues and cutting null rate from 34% to <2%, enabling the team’s first automated reporting dashboard.

I’ve also worked on research-grade data engineering: I built an RDF compression pipeline (OpenRefine + custom OWL ontology) that reduced a 77 GB dataset to 9 GB (88%), cutting SPARQL query latency by ~60% for real-time analytics on 120 years of Olympic records. I contributed to an Olympics Semantic Web portal using SPARQL and designed ontology relationships in Protégé.

Across projects, I combine strong analytics with explainable AI and modern NLP/RAG. I built a Multi-Disease Prediction System with XAI (SHAP/LIME) and a full-stack healthcare portal, plus a RAG-based teaching assistant that uses Whisper transcription, timestamped chunking, embedding retrieval, and LLM inference with a Streamlit UI.