Kiefer Land

Kiefer Land

AI Quality & Evals Engineer

As agents write more of the code, the scarce resource becomes verified trust, and someone has to build the machinery that produces it.

verified
3Years in software testing & QA
7Security certifications
0QA-attributed rollbacks

Track record

2026

Scaling QA throughput with Claude Code agentic workflows — custom /ship and /loop skills and MCP integrations, including an eval-graded custom skill scoring 100% vs. a 61.9% baseline. Grew an internal Next.js/Supabase app's test suite from 316 to 940+ tests at 100% line+branch coverage on changed files, hardened with Stryker mutation testing (80% → ~94%), plus a live-database RLS integration suite. Shipped 6 side projects this year using autonomous agent pipelines with full CI gate discipline (+53K verified lines merged in a single month).

2025

Built an AI-assisted test case generation system in ChatGPT with human oversight for requirements analysis, test planning, and scenario authoring — improving test coverage 35%. Used GitHub Copilot and VS Code Agent Mode to cut regression cycle time 50%.

2024

Promoted to Software QA Engineer I at TCWGlobal — expanded API automation coverage 30% → 70% (Postman/Newman, Jenkins CI) and E2E automation 20% → 80% (Playwright). Earned an M.S. in Cybersecurity & Information Assurance from Western Governors University.

2023

Promoted to Software QA Analyst at TCWGlobal (Oct) — authored 150+ functional and edge-case regression scenarios and led QA through a PHP → Next.js/NestJS migration with zero rollbacks. Earned a B.S. in Cybersecurity & Information Assurance from Western Governors University.

2022

Started at TCWGlobal as an IT Support Technician (Feb) — built an IT asset management system (70% better inventory accuracy) and managed the lifecycle of 150+ devices with NIST 800-88–compliant data disposal. Promoted to Junior System Administrator (Dec) — automated MacBook provisioning with Jamf (60% faster new-hire setup) and led security hardening across 250+ Windows/Linux systems, sustaining 99%+ uptime.

Working principle
Trust doesn't scale; evidence does. Every claim this site makes about my work is meant to be something you could, in principle, replay and check.