Data Analytics, Business Intelligence, and Automation Systems
I build practical analytics and automation systems for operations teams: dashboards, reporting workflows, request trackers, searchable knowledge tools, and automations that reduce manual follow-up.
Each project is written as a case study with the problem, system design, data flow, implementation choices, and evidence.
Live US and European electricity analytics platform using scheduled API ingestion, time-series modeling, anomaly scoring, forecasting, source freshness, and dashboard delivery.
Value: turns scattered public grid feeds into one live view of demand, forecasts, and anomalies across 65+ US balancing authorities and 20 European zones.
Live demo available. Public-data project with real screenshots and architecture notes.
Hypothesis testing and statistical inference platform on CFPB HMDA mortgage data. Five preregistered hypotheses, parametric and non-parametric and permutation and bootstrap and Bayesian sensitivity validation, BH-FDR and Bonferroni correction, effect sizes with confidence intervals, and explicit causal framing caveats.
Value: turns 210,643 raw HMDA records into a fair-lending screening dashboard where every finding is validated by parametric, non-parametric, and resampling methods and corrected for multiplicity across the test family.
Live demo available. Public-data project with current screenshots, multi-method validation, and explicit limitation framing.
Production-style factory analytics platform using a seeded synthetic assembly dataset, PostgreSQL star schema, FastAPI endpoints, validation checks, and a Next.js executive dashboard.
Value: turns raw operational logs into an executive health dashboard: 90.3% plant OEE with a defensible path from source records to decisions.
Live demo available. Synthetic-data project with current screenshots, validation evidence, and full source repo.
Bring-your-own-key text-to-SQL agent for the synthetic factory warehouse, with client-side model calls, schema grounding, read-only SQL guardrails, and visible query evidence.
Value: lets anyone query the warehouse in plain English with zero LLM keys held on the backend and read-only SQL guardrails on every generated query.
Live demo available. Synthetic-data AI project with current screenshots, safety guardrails, and full source repo.
Explainable Ohio public drinking water screening model using public data, transparent scoring, service-area boundaries, source-water overlays, API-backed search, and dashboard delivery.
Value: converts 16,339 Ohio water-system records into a transparent, validated risk-screening tool (22/22 checks passed) that states its limits clearly.
Live demo available. Public-data project with current screenshots, 22 validation checks, and methodology notes.
Hybrid arXiv paper recommender on a 28,000-paper OpenAlex snapshot of Computer Science papers. Compares popularity, TF-IDF, a MiniLM sentence-transformer, citation-graph ALS, and a learned blend, with bootstrap-CI evaluation and FAISS ANN serving.
Value: turns the OpenAlex citation graph and abstract corpus into a paper recommender that beats the popularity baseline by roughly 20× on MAP@10 and serves top-10 in under 100 ms p95.
Live demo available. Public-data project with current screenshots, bootstrap-CI metric leaderboard, and full source repo.
Hybrid book recommendation system on the Goodbooks-10k corpus combining popularity, TF-IDF content, implicit-feedback ALS, a neural two-tower, and a learned blend, with held-out evaluation, FAISS ANN serving, and a side-by-side interactive demo.
Value: turns 5.9M Goodreads-style ratings into a hybrid recommender that beats the popularity baseline by ~61× on MAP@10 and serves top-10 results in under 100 ms p95.
Live demo available. Public-data project with current screenshots, bootstrap-CI metric leaderboard, and full source repo.
Natural-language aviation safety analytics over NTSB final reports, combining guarded SQL, cited narrative retrieval, validated charts, visible SQL, and audit trails.
Value: answers aviation-safety questions over 7,462 NTSB reports while keeping exact SQL counts and cited narrative evidence separate and auditable.
Live demo available. Public-data project with current screenshots, source citations, SQL evidence, and evaluation notes.
I build analytics and automation systems that make operations easier to understand, easier to track, and easier to improve.
My work focuses on the full path from messy process to usable system: intake, data structure, workflow, reporting, and adoption. A dashboard is only useful when the inputs are trustworthy and the process behind it is something people can actually follow.
This portfolio is written for recruiters, hiring managers, and technical interviewers who want to see how I solve real problems beyond a list of tools. Each case study explains the problem, solution, architecture, tradeoffs, and results.