Results for “measured-boot”

Read-only audit of product-analytics instrumentation: event taxonomy, funnel completeness, consent-gated firing, dead/duplicate/phantom events. Use when "audit our analytics", "are we tracking the right events", "funnel instrumentation", or before iterate-post-launch. PostHog/Amplitude/Mixpanel/GA4.

Build monitoring dashboards that answer real operator questions for Grafana, SigNoz, and similar platforms, turning metrics into actionable operational views.

Read-only audit of a codemod or bulk mechanical transform for behavior-preservation — compiles/lints is not same-behavior. Use when "did this codemod break anything", "audit this bulk refactor", or before merging a mass find-replace. Diff quality → audit-code-review. SQL → plan-data-integrity.

Audit server-side paid-media measurement, including tag management, conversion APIs, event taxonomy, deduplication, consent, hashing, data quality, and observability.

Builds statistical plots with the seaborn Python library and pandas DataFrame integration, on attractive matplotlib-based defaults. Use for quick exploration of distributions, relationships, and categorical comparisons — box plots, violin plots, swarm/strip plots, KDE/histograms, pair plots, joint plots, regression plots, correlation heatmaps, and faceted small multiples (relplot/displot/catplot/lmplot). For interactive/hover/zoom charts defer to alterlab-plotly; for exact journal/manuscript styling (column widths, point fonts, CMYK, vector export) defer to alterlab-scientific-viz; for low-level custom matplotlib figures defer to alterlab-matplotlib (seaborn integrates with it for fine-tuning). Part of the AlterLab Academic Skills suite.

Design, audit, and improve analytics tracking systems that produce reliable, decision-ready data.

Computes the SQuAD metric using torchmetrics, given predictions and ground truth. Use when evaluating question-answering outputs with exact match and F1 scores.