Vibe-coded click-the-state quiz, deployed on Vercel. Click anywhere to try it.
Public Safety ETL Pipeline on Azure2024
Azure medallion pipeline ingesting open data APIs into a Bronze–Silver–Gold lake. Databricks ETL with Power BI dashboards for public safety analytics.
MQTT Smart Meter Data Streaming Pipeline2025
Multi-meter monitoring for Gen5 Riva Itron smart meters at Xcel Energy. Replaces manual validation with MQTT telemetry, InfluxDB, and Grafana dashboards.
Interactive LLM Data Visualization Assistant2025
Graduate research using Ollama-hosted LLMs for interactive analytics. Prompt-engineered JSON output rendered as Plotly charts with multi-turn context.
Azure Databricks: Cleaning Energy Meter ASCII Data2024
PySpark and Azure Databricks pipeline for irregular ASCII meter files. Produces normalized Parquet and CSV datasets for analytics.
Public Safety ETL Workflow with Dagster2026
Dagster pipeline that pulls Minneapolis police incidents from the Open Data API, cleans them, and writes Parquet. Assets flow raw → processed → enriched with community and Metro Council place joins on a 6-hour schedule.
MQTT-Based IoT Telemetry Pipeline2026
Automated DS18B20 temperature collection on Raspberry Pi with live Grafana charts. MQTT and InfluxDB for live data; Django WebApp archives CSV exports to PostgreSQL.
Airbnb Data Preprocessing & Feature Engineering for ML2024
End-to-end scikit-learn pipeline for Airbnb price prediction. Includes EDA, feature engineering, and ensemble model comparison.
YouTube Metadata Ingestion Pipeline2023
Ingests Seattle Data Guy video metadata from the YouTube Data API using Python. Extracts and transforms video metadata into a structured dataset, with automated Excel export for downstream tracking and analysis.
Unstructured Data Parsing & Vector Search2023
Django search engine with NLTK preprocessing and term-document indexing. Relevance-ranked retrieval backed by PostgreSQL.
Scientific Text Preprocessing & NLP with SciBERT2022
M.Sc. thesis: fine-tuned SciBERT and BioBERT on annotated SoMeSci data. Classifies software purpose mentions in scientific literature.
Regex-Based Text Parsing & Extraction Pipeline2021
Python pipeline mining 3.2M+ PubMed and PLOS XML articles. Extracts and analyzes Research Resource Identifier (RRID) adoption.
