Architectures

AI coding agent runtime — 512K lines of TypeScript. Query engine, permission-gated tool system, multi-agent orchestrator, bash security parser, memory subsystem, and unreleased autonomous daemon.

Autonomous LLM pretraining research agent by Andrej Karpathy — an AI agent that modifies code, trains models, evaluates results, and iterates indefinitely while you sleep.

Batteries-included agent harness built on LangGraph — planning, filesystem, sub-agents, and context management out of the box.

Workflow automation platform with 300+ integrations, native AI agent capabilities, and fair-code license — v2.14.0 monorepo (pnpm + Turborepo)

OpenClaw plugin for NVIDIA OpenShell — sandboxed agent execution with managed inference and declarative policy enforcement

Personal AI assistant gateway — 82 channel plugins, 40+ LLM providers, file-based sessions, extensible plugin SDK with 60+ subpath modules.

The Agent Operating System — 14 Rust crates, 137K LOC, 16 security layers, single ~32MB binary with true kernel architecture.

Control plane for AI agent orchestration — never runs AI directly. Task queue + org chart + audit log + budget tracker that orchestrates external agent runtimes via adapters.