Learning
is the pinnacle
of human
existence.

A 5,000+ course catalog with a multi-year QA backlog and only 40% audit coverage.

Architected an automated evaluation pipeline using Codex, Claude, and Python — graded against WCAG 2.1 AA and instructional-design rubrics, surfaced findings to reviewers.

Manual review cycles cut by 65%. Audit coverage expanded to 100%. Backlog eliminated.

Course delivery taking ~6 months per cohort across the org.

Built an AI-assisted authoring tool with embedded WCAG and ID-standards checks, plumbed into Canvas and faculty workflows.

Delivery cycle halved. 100+ courses shipped in 2–3 months across EdPlus.

Review → fix → verify cycles consumed thousands of labor hours each year.

End-to-end pipeline: automated audit findings flow into remediation queues with verification checks on the way out.

Manual cycle reduced 70%. ~5,000+ labor hours saved annually across 800+ courses.

Amazon Delivery Central Operations had no technical infrastructure for Kirkpatrick L3/L4 measurement — costly retraining cycles, anecdotal decisions.

Architected an end-to-end evaluation system on Python + Slack webhooks + Asana workflow + Amazon Quick Suite, bypassing VBA limits for robust data collection and AI-powered analysis.

Survey participation +56% (44.49% → 69.17%). Pinpointed 20% outdated SOPs. ~$50K saved annually; 80+ hours of manual collection eliminated per teammate per quarter.

Static training material wasn't producing strong first-attempt outcomes on critical content.

Built variable-based branching simulations in Articulate 360 with AI-driven scenario logic.

5/5 facilitator feedback. Trained 300+ associates with on-time delivery across all L&D submissions.