Builder · Autonomous AJ
I build agent fleets that ship real work.
I'm Andrew J. Pyle — a full-stack engineer & AI-native operator. I built AACC, an autonomous command center that runs AI agents as real teammates: they ship SEO, analytics, content, and code across a portfolio of dozens of live sites. Everything below is what that engine builds.
One operator. 35 agents across 31 integrations ship code, SEO, content, and whole sites. A lead agent routes the work. It asks before anything is sent, paid, or deleted.
AndrewassignsAdversarial plan review — a red-team panel grounded in your own past mistakes.
Agent-memory hygiene — finds drift, contradictions, and stale claims across your memory.
Test-coverage swarm with a mutation gate, so it can't write assertion-free tests.
See what it builds↓
Curiosity domains
The range, and what backs it.
Curious about a lot of things. Building is how I make sense of them, and the fleet is how that curiosity turns into shipped software.
AI & agents
40 live · 5 shown
Hover to explore · Click to navigate
Selected work
Curiosity that shipped.
AACC — Autonomous AJ Command Center
The agentic command center behind the whole portfolio. It dispatches AI agents to ship SEO, analytics, content, and CI across dozens of sites — with a human-in-the-loop approval queue. The operating system for everything else here.
USA Prep Football
A multi-tenant high-school-football platform templated for all 50 states — real-time scores, previews, coach profiles, and team analysis, run as one network.
OZ Intelligence
A real-time sports-data platform and public REST API — live scores, standings, and historical records fed by automated pipelines, served as clean JSON.
The Sports Machine Network
A family of per-sport analytics sites spanning pro and college baseball, basketball, football, and hockey.
The War Atlas
An operational atlas of military history — a source-cited catalogue of wars, battles, commanders, and factions, with a map + timeline layer.
Civics Matter
A reference platform for U.S. government — Congress, state legislators, counties, and districts — with plain-language explainers of how the system works.
The whole network
103 live domains, one operator.
5 multi-site networks plus 27 standalone builds — every one live, all run by the same engine. Not the curated highlights; the full roster.
What I ship
AI skills & open source.
The tools behind the portfolio — free libraries, installable agent skills, and products in the works.
AI-Visibility
SoonRank-tracking for LLMs — see where your brand shows up across the answer engines.
content_gate
SoonWill this content hurt your AI-search visibility? A do-no-harm SEO / AEO / GEO gate.
aacc-spar
Adversarial plan review — a red-team panel grounded in your own past mistakes.
aacc-substrate-audit
Agent-memory hygiene — finds drift, contradictions, and stale claims across your memory.
aacc-coverage-swarm
Test-coverage swarm with a mutation gate, so it can't write assertion-free tests.
ailab-core
Shared core for the ailab-* discipline labs: provider seam, cassette replay, and the regression gate primitive.
ailab-rag
Inspectable RAG lab: chunk, retrieve, rank behind an honest, saturation-guarded eval gate with a real recorded-model pass. Stdlib-only.
ailab-prompting
Prompt-engineering lab: zero-shot vs few-shot vs chain-of-thought plus a prompt-injection defense, behind an honest eval gate. Stdlib-only.
ailab-agents
Bounded multi-hop QA agent: tools + context + MCP + evals. Fourth ailab-* AI-discipline lab, built on ailab-core.
ailab-evals
LLM evals harness: datasets, runners, calibrated LLM-as-judge, pairwise mode, and a regression-gate CLI. Stdlib-only.
ailab-template-python
Template for Python AI-discipline labs: uv, pytest, ruff, eval-gated CI, secret-scan wall, non-root Docker.
ailab-template-dotnet
Template for C#/.NET 9 AI lab sidecars: Minimal API, Native AOT, xUnit, health probes, eval-gated CI, secret-scan wall.
google_search_console_mcp
MCP server for Google Search Console — search analytics, URL inspection, and sitemap management for AI assistants (Claude Desktop/Code). Real API calls only.
programmatic_pages
Render real, static HTML from a Django queryset — schema.org + canonical, one page per genuine record.
url_indexing_playbook
Ping search engines about your new URLs. Django management commands for GSC + IndexNow + Cloudflare Crawler Hints.
consent_mode_v2
Google Consent Mode v2 — dependency-free drop-ins for Next.js, Vite/React, and server-rendered sites. No build step.
agent_action_approvals
A durable, human-in-the-loop approval queue for agent-initiated actions, as a small Django app.
content_homogeneity_gate
A domain-agnostic gate that catches templated, near-duplicate content before you publish it at scale.
cc_insights_scope
Repo-scoped, cross-machine Claude Code /insights — analyze one project's sessions instead of every local session.
agent_db_guard
A config-driven Claude Code hook that stops an AI agent from running a destructive command against a production database by accident.
ai-exec-org-template
An AI executive team that cannot act: 9 seats read your real numbers, one engine writes one draft brief with today's one action, two dead-man alarms. Zero dependencies. Built for Claude Code.
Have something you need built or fixed?
I build production Django / Next.js platforms and human-supervised AI-agent systems. Solo, senior, and fast. Tell me what you are building.