AI Enablement Engineer | Operational Intelligence | DevTools Python · Rust · TypeScript · Linux Systems · Full CI/CD
I apply lean manufacturing systems thinking to AI and software engineering. 17 years in enterprise operations (Toyota, IBM) before I touched a model. The result: systems that ship reliably, not just demo well.
AI workflow orchestration — multi-agent pipelines with cost tracking, checkpointing, and structured feedback loops Document intelligence — investigative-grade ingestion, entity extraction, and co-occurrence analysis Developer tooling — CLI utilities for AI teams: spend tracking, agent auditing, context generation Production AI applications — autonomous content pipelines running at scale across multiple languages
Animus — Multi-Agent Orchestration Framework
Coordinate specialized AI agents across enterprise workflows with budget controls, quality gates, and checkpoint/resume. FastAPI backend · React dashboard · Quorum coordination protocol (PyPI) · 13,100+ tests
BenchGoblins — AI Fantasy Sports Decision Engine
League-centric start/sit, trade analysis, and matchup projections. Live on Fly.io + Vercel. Next.js · FastAPI · PostgreSQL · Redis · Stripe · 1,988 tests · 99% coverage
DOSSIER — Document Intelligence Platform
Investigative-grade document analysis: ingest PDFs/emails/scans, extract named entities, surface co-occurrence relationships. SQLite FTS5 · Custom NER · Auto-classification · 1,055 tests · 97% coverage
agent-lint — Agent Workflow Auditor
CLI tool for cost estimation and anti-pattern detection in agent workflow YAML configs. PyPI Static analysis · 17 lint rules · CI integration · Zero LLM dependency
Argus Overview — Linux Multi-Window Manager for EVE Online
Real-time 30 FPS window previews, team management, auto-tiling layouts, EVE settings sync. 2,500+ downloads · 1,892 tests · 96% coverage
| Tool | What | Tests |
|---|---|---|
| promptctl | Claude API toolkit — prompt engineering, code review, doc intelligence | 286 |
| memboot | Zero-infra persistent memory for LLMs — SQLite + TF-IDF (PyPI) | 285 |
| claudemd-forge | CLAUDE.md generator/auditor for AI coding agents (PyPI) | 569 |
| ai-spend | AI API cost aggregator — Anthropic + OpenAI billing in one view | 222 |
| mcp-manager | MCP server manager — discover, monitor, manage across IDEs | 139 |
83 workflow files across 20 repos. 0 open code scanning alerts.
| Capability | Evidence |
|---|---|
| Containerization | Multi-stage Dockerfiles, docker-compose, templates |
| Kubernetes | Deployments, HPA, Ingress, Kustomize, Helm chart |
| CI Pipelines | Lint → test → SAST → build → deploy across Python, Rust, TypeScript |
| Security Scanning | CodeQL + gitleaks + Bandit + Semgrep + Trivy + pip-audit |
| Live Deployments | BenchGoblins on Fly.io + Vercel, 4 packages on PyPI via OIDC |
| Benchmarks | Automated performance regression tracking on gh-pages (3 repos) |
| Metric | |
|---|---|
| Tests across active repos | 30,700+ |
| Projects with CI green | 20 |
| Live deployments | BenchGoblins (Fly.io + Vercel), 4 packages on PyPI |
| Security posture | 0 code scanning alerts across all repos |
| Real users | Argus Overview: 2,500+ downloads |
| Domain | Tools |
|---|---|
| Languages | Python, Rust, TypeScript |
| AI/ML | Claude API, OpenAI, Ollama, multi-agent orchestration |
| Backend | FastAPI, SQLite, PostgreSQL, Redis |
| Frontend | React, Next.js, React Native/Expo, PySide6 |
| Infrastructure | Docker, Kubernetes, Helm, GitHub Actions, Fly.io, Vercel |
| Security | CodeQL, Bandit, Semgrep, Trivy, gitleaks, SAST pipelines |
Operational reliability over algorithmic elegance. A system that ships predictably beats a system that demos brilliantly. Standard work, visual management, poka-yoke — applied to software.