Data Inspector HXD

Data Inspector HXD analyzes datasets to reveal structure, quality, and core statistics without modifying the source data. It supports CSV, JSON, and Parquet formats and is intended for data engineers, data scientists, and QA analysts during ingestion, validation, feature engineering, and model readiness. Users gain visibility into column types, missing values, and value distributions, enabling rapid data understanding, anomaly detection, and governance documentation. The tool emphasizes reproducibility and auditability to support cross-team reviews and compliance requirements.
How it works conceptually: the engine infers schemas, computes per-column statistics (min, max, mean, median, std for numeric; unique counts and top values for categorical), and flags missing or suspicious data patterns. Advanced options compute correlation matrices, percentile distributions, and optional outlier indicators. For large datasets, profiling can sample data or operate incrementally, returning compact previews and exportable reports in JSON or CSV.
Use cases and differentiators: quality checks before modeling, data cataloging, feature understanding, and compliance reporting. Unlike full ETL tools, it remains read-only and focused on inspection, providing precise metrics rather than transformations. It supports locale-aware numeric parsing and ISO date formats, aiding teams across regions. Ideal for teams needing quick, auditable data profiles to accelerate data-driven decisions.