data-hygiene

Test-contamination detector - finds train/eval overlap via deterministic shingle matching, per-source rollups, CI-ready exit codes. Standard library only.

A dependency-free, no_std Rust crate for secure in-memory secret handling, constant-time checks, and automated clear-on-drop zeroization.

One-page reference on GTIN, GDSN, item-setup, and freight-class rules for specialty food operators, with a downloadable PDF.

Python CLI that profiles every column in an Excel file and flags hidden data-quality problems (mixed types, sentinel values, missing data, cardinality anomalies) in plain English. Outputs PDF, HTML, JSON, or annotated-Excel reports, with strict cell-level type detection that catches mixed-type columns pandas silently coerces.

Product data validation tool for specialty food brands preparing for national retail

Contact data hygiene CLI for music PR. Email validation, deduplication, and enrichment.

TRACER: audit-first intrusion-detection benchmarking toolkit with time-robust evaluation and leakage diagnostics.

Audits messy Excel workbooks for the data-quality issues that show up in real consulting engagements: mixed formats, misused fields, placeholder floods, and phantom duplicates. Generates HTML, Excel, and PDF reports.

Dimension and weight validation for product data — catches the dim-weight defects behind freight chargebacks and compliance fines. Python.

USADATA, Inc. — independent third-party profile of a public API surface, by API Evangelist. USADATA is a data and marketing intelligence company founded in 1998 and headquartered in New York City, backed by Insight Partners. It maintains a database of 260+ million US consumer records, 15+ million businesses, and 190 million households, updated mont

A complete product data health audit for a fictional specialty food company. Five artifacts (HTML report, executive tearsheet, dashboard, Excel workbook, Shiny calculator) from one R+Quarto pipeline. Portfolio piece.

Salesforce-native data-hygiene platform that finds and fixes the CRM data issues that break Agentforce agents: auto-fixing safe changes, routing risky ones for approval, and escalating the rest, via a headless engine, an admin app, and a chat agent.

Local web app that anonymizes tabular data through guided column-by-column review. Deterministic mappings, format-preserving fakes, bidirectional reverse lookup.

Compare two tabular data files and get a clear report of what changed — rows added, removed, or modified, with before/after values.