Computer Science Master's graduate (NJIT) who builds end-to-end data and AI solutions — from scalable ETL pipelines and big-data processing to machine learning, NLP, and Generative AI. IEEE-published, with deployed projects across the data stack.
I work end-to-end across the data lifecycle — designing ETL/ELT pipelines, building and deploying machine-learning and LLM applications, and turning messy data into reliable, production-ready products. I care about clean code, honest evaluation, and solutions that actually ship. Currently going deeper on agentic AI, Snowflake, dbt, and Airflow.
The tools I reach for across data engineering, data science, and ML/AI.
Selected work — each links to its GitHub repository.
End-to-end fraud models on imbalanced data; best model selected by ROC-AUC/PR-AUC and deployed as a real-time FastAPI scoring service.
View repo ↗LLM app with retrieval-augmented generation, sentence-transformer embeddings, and grounded answers with source citations.
View repo ↗Distributed batch processing on Hadoop/HDFS with a Combiner for local pre-aggregation; deployable on AWS EMR.
View repo ↗Distributed pipeline using SQS to decouple parallel workers running Rekognition object detection, results to S3.
View repo ↗TF-IDF + voting/stacking ensembles for toxic-comment classification. Basis of a co-authored IEEE publication.
View repo ↗Benchmarks RF, SVM, Gradient Boosting & a neural net with cross-validation and automatic best-model selection.
View repo ↗Normalized 3NF schema, indexed analytical SQL, and an interactive Streamlit reporting dashboard.
View repo ↗Automated rvest/xml2 scraping with tidy cleaning and ggplot2 visualizations.
View repo ↗Open to Data Engineering, Data Science, ML/AI, and Software Engineering roles. The fastest way to reach me is email.