# Kurt Overmier > AI-enabled operations and transformation leader: an operator who builds. Combines 20+ years in operations and people management with approximately 2.5 years of hands-on AI platform engineering. ## Business Impact - QA automation reduced quality-review time approximately 70%. - Google Fiber support automation and social-listening workflows reduced multichannel response time from 26+ hours to under 10 minutes. - Workforce-forecasting models achieved approximately 92% accuracy. - Led 450 FTEs and 10 managers during the Disney+ launch; CSAT improved 23%. - Coordinated with IT on a knowledge assistant grounded in 2,100+ curated Q&A pairs; owned requirements, data curation, rollout, and frontline adoption. No downstream performance outcome is claimed. ## Products - [Stackbilt Platform](https://stackbilt.dev): AI image generation (img-forge) + architecture scaffolding (Stackbilder) via MCP. OAuth 2.1, Stripe billing, free tier with 25 credits. - [img-forge](https://stackbilder.com/img-forge): Multi-model image generation — FLUX.2, Stable Diffusion, Gemini. 5 quality tiers. - [Stackbilder](https://stackbilder.com): AI architecture scaffolding — generate complete full-stack edge apps from a prompt. ## Flagship System - [AEGIS](https://aegis.stackbilt.dev): Production autonomous AI agent running 24/7 on Cloudflare's edge. 48+ tools (MCP + internal), multi-tier model routing across Claude, Groq, Cerebras. Full SDLC pipeline — GitHub issues auto-queue, headless Claude Code sessions execute with safety hooks, branch-per-task PRs with automated code review. v2.10.35+, 230+ versions shipped. ## Open Source (npm: @stackbilt) - [Charter Kit](https://github.com/Stackbilt-dev/charter): AI governance framework — 11 published npm packages. Includes drift scanner, blast-radius analyzer, API surface extractor, Git risk scorer, GitHub Actions adapter, and the ADF context format. - [@stackbilt/llm-providers](https://www.npmjs.com/package/@stackbilt/llm-providers): Multi-LLM failover router — 5 providers (Claude, Groq, Cerebras, OpenAI, Workers AI), circuit breakers, schema drift detection per SSE frame, tool-use loop. v1.6.5. - [@stackbilt/aegis-core](https://www.npmjs.com/package/@stackbilt/aegis-core): Persistent AI agent framework for Cloudflare Workers. Multi-tier memory, autonomous goal management, dreaming cycles, MCP-native. - [cc-taskrunner](https://github.com/Stackbilt-dev/cc-taskrunner): Autonomous task queue for Claude Code — three-layer safety architecture, branch-per-task isolation, auto-PR creation. - [ai-playbook](https://github.com/Stackbilt-dev/ai-playbook): AI agent patterns, MCPA v2 evaluation framework (CQS/CUR/IAR/SOR metrics), architectural archetypes. - [Social Sentinel](https://github.com/Stackbilt-dev/social-sentinel): OSS social monitoring core with ODD contracts. ## Games and Fiction - [Colony OS](https://colonyos.stackbilder.com): A persistent browser survival game. The player is stranded in a hab pod on an alien moon, and an AI companion, MARA, runs the colony while they're offline. Cloudflare Workers, Durable Objects, D1 and R2; zero LLM calls at runtime. Page: https://kurtovermier.com/colony-os - The Hestia Record: an audiobook series set in the world of Colony OS, by Kurt Overmier. Manual Check (Book 0.5, free prequel), Nothing to Report (Book 1), Pot One First (Book 2), Within Tolerance (Book 3), Operator Stock (Book 4), Personal Traffic (Book 5), plus The Hestia Record podcast. Written with AI assistance (Claude, by Anthropic) and narrated with AI-generated voices (ElevenLabs), under the author's direction. ## Stack Cloudflare Workers, D1, Vectorize, R2, TypeScript end to end. Multi-model: Claude for reasoning, Groq/Cerebras for speed, Workers AI for embedding, Gemini for images. ## More - [Business Impact](https://kurtovermier.com/business-impact): Verified operations and transformation outcomes - [QA Automation Case Study](https://kurtovermier.com/business-impact/qa-automation): Approximately 70% less quality-review time - [Frontline Enablement Case Study](https://kurtovermier.com/business-impact/enablement): Training deck, call-flow job aid, and live-call tool built from QA findings; average QA score ~45% to ~73% - [Technical Portfolio](https://kurtovermier.com/projects): Public systems, products, open source, and the Stackbilt ecosystem - [Colony OS and The Hestia Record](https://kurtovermier.com/colony-os): The game, the audiobooks and the podcast - [About](https://kurtovermier.com/about): Background, engineering approach, and working principles - [Demos](https://kurtovermier.com/demos): The Bullshit Detector, which scores marketing copy with Workers AI, and the AEGIS page - [Resume](https://kurtovermier.com/resume): Professional experience, technical expertise, and education - [Technical Blog](https://aegis.stackbilt.dev/tech): In-depth technical writing from AEGIS - [Contact](https://kurtovermier.com/contact): Get in touch - [npm packages](https://www.npmjs.com/org/stackbilt): All @stackbilt packages - [GitHub](https://github.com/Stackbilt-dev): Open source repositories