Walmart Data Engineering Project
An end-to-end TPCH orders pipeline with staging models, fact tables, data marts, reusable macros, and data-quality tests orchestrated through Airflow DAGs.
Data engineer · Baltimore, MD
Hi, I’m Aryan.
I build data pipelines, cloud platforms, and machine learning tools. I recently completed my M.S. in Information Systems at UMBC, where I worked on geospatial data systems and AI research.
Open to data engineering opportunities and relocation.
About me
I’m Aryan, a data engineer based in Baltimore, with an M.S. in Information Systems from the University of Maryland, Baltimore County (August 2026, GPA 3.95) and a B.E. in Computer Science.
My work spans ELT pipelines, warehouse modeling, and cloud data platforms. I’ve shipped a geospatial platform used by international research teams, built reproducible biomedical data workflows, and placed second nationally in the NSF HDR coastal flooding challenge.
I like working through the practical parts of a project: cleaning the data, building the pipeline, and checking whether the results are useful. I’ve also taught programming at iCode and Johns Hopkins CTY.
Modern data stack
dbt · Snowflake · Airflow
SQL · Python · Spark
Data modeling · Data quality
Cloud & platforms
GCP · Docker · Git · CI/CD
FastAPI · Flask · REST APIs
PostgreSQL · MySQL · SQLite
Analytics & research
Power BI · DAX · Plotly · R
Pandas · XGBoost · LLMs
FAIR data · BioCompute Objects
2024–2026 · Graduate study
M.S. Information Systems
University of Maryland, Baltimore County · August 2026 · GPA 3.95
2020–2024 · Undergraduate study
B.E. Computer Science
Shah & Anchor Kutchhi Engineering College · Mumbai, India · May 2024
Selected projects
Explore all projects ↗Data Engineering & Analytics
Cloud data platforms, warehouse models, and dashboards that make complex data usable.
Geospatial Research Platform
A deployed platform for international research teams: hex-level GeoTIFF classification, annotation assignments, QC review queues, and satellite-data export pipelines.
Live Market Data Dashboard
Ingests live Excel market exports, cleans malformed input, and serves a JSON API and browser dashboard deployed on Render.
AI & Machine Learning
Forecasting, retrieval systems, and human-centered evaluation—from experiments to applied models.
Coastal Flood Forecasting Pipeline
Transformed 70 years of hourly tide-gauge records into features for 14-day forecasting. Achieved ROC-AUC 0.871 and leaderboard F1 0.94 in the NSF HDR Challenge.
LLMs for Dialogic Reading
Reproducible Python/Jupyter evaluations of GPT-4, Gemini, and Llama with tracked prompt versions, model configurations, and versioned results.
PDF-to-RAG Ingestion Pipeline
Structure-aware PDF-to-Markdown ingestion, hierarchical chunking, embedding generation, and a persistent vector store with CLI ingestion and retrieval workflows.
Experience
Building production platforms, reproducible pipelines, and research tools has shaped how I approach data engineering.
View full experience ↗Data & Research Engineer · iHARP Fellow
University of Maryland, Baltimore County
Architected and shipped a FastAPI geospatial platform for international research teams, integrating GCP object storage, ArcGIS basemaps, and a SQLite workflow layer. Built satellite-data ingestion and export pipelines with annotation tracking and QC review, plus Power BI semantic models and dashboards for the UNDP-aligned Nature Relationship Index.
Machine Learning & Data Engineer
GWU HIVE Lab (Mazumder Lab) · Remote
Built extraction and harmonization pipelines for public multi-omics and PMID-sourced datasets, producing modeling-ready feature tables with automated QC. Containerized analysis workflows with Docker and documented them as BioCompute Objects for FAIR-compliant reproducibility and lab reuse.
Graduate Researcher · M.S. Thesis
UMBC · Advisor: Prof. Karen Chen
Designed reproducible Python/Jupyter pipelines benchmarking GPT-4, Gemini, and Llama. Tracked prompt versions, model configurations, and result sets to support structured experiments and human-centered evaluation of dialogic reading questions.
Teaching Assistant
Johns Hopkins Center for Talented Youth
Supported advanced STEM programming, guided student projects, and provided feedback on machine learning and data science assignments.
Research
More about my research ↗Master’s thesis · UMBC
Evaluating Large Language Models as Generators and Judges of Dialogic Reading Questions for Early Literacy
Investigates large language models as both generators and evaluators of dialogic reading questions for early literacy. The research examines question quality and instructional alignment for kindergarten and first-grade learners.
Aryan Dhiren Jagani · University of Maryland, Baltimore County · ProQuest Dissertations & Theses, 2026 · Publication no. 32792284.
Conference paper · ICSSAS
CogniPoseAI: A Futuristic AI-Enhanced Personal Trainer
Using computer vision and pose estimation to analyze exercise posture, count repetitions, and provide real-time feedback.