Adventure Wave Labs
Building the future of agentic engineering, AI-powered automation and open source tooling
Our Labs
Two research labs, one mission — pushing the boundaries of agentic AI, open source, and developer experience across the Americas.
Agentic Engineering & Platform Innovation
github.com/adventurewave-labsA Los Angeles-based research lab focused on agentic AI, platform engineering, and developer tooling. AWLabs created Turbo Flow and now builds its successor Turbo Rig — the governed agentic coding rig — plus rig-lite, the open-source governance kit extracted from it.
- Creator of Turbo Flow & Turbo Rig (175 GitHub stars)
- Agentic Platform Engineering research
- Open source Rust & Shell tooling
- Multi-agent AI orchestration systems
Our Latin American affiliate lab driving AI adoption and education across the region. CMLabs specializes in bilingual AI tooling, community-driven development, and making cutting-edge agentic technology accessible to Spanish-speaking developers worldwide. CMLabs was a key contributor to Turbo Flow en Español (2025–2026), bringing the agentic development platform to the Latin American developer community; today it localizes the lab work into Spanish.
- Latin American AI education & adoption
- Bilingual developer tooling (EN/ES)
- Community-driven open source projects
- AI consulting & automation services
Lab Demos
Everything we build ships live. Explore the products, tools, and sites running in production across both labs.
Core Initiatives
Our flagship products and research programs — the driving force behind both labs, powering the future of agentic engineering and platform innovation.
From autonomous AI workflows to golden-path developer platforms, these are the initiatives defining the next era of engineering.
Turbo Flow v5.2 — the rules layer. One product: rig-lite, a portable governance kit (a fail-closed cross-model review gate, a written constitution, git-versioned memory) that works with any harness in any repo. Versions 1–4 (2025–2026) were the agentic development environment — 215+ MCP tools, 60+ agents at the v4 peak; that era is preserved in the version tags and led to Turbo Rig.
Turbo Rig v5 — the successor to Turbo Flow. A rig that orchestrates AI coding agents instead of replacing them: a cross-family code-review gate where a different model family reviews every diff, git-versioned agent memory, and an AGENTS.md constitution. 8 thin scripts in bash + Python, zero daemons, human-only merges. The gate applies a published finding: in a controlled study of 116 tasks (Xiang et al., arXiv:2607.21656), cross-model review lifted pass rates from 71.6% to 89.7%, outperforming same-family self-review.
Agentic Platform Engineering
Core Research Initiative
Building autonomous, AI-driven platform engineering systems that transform how developers build, deploy, and manage infrastructure. From multi-agent orchestration to golden-path developer onboarding.
By The Numbers
Totals across our GitHub orgs, Vercel, and crates.io — live where the APIs allow.
121
GitHub Repos
234
Total Stars
48
Followers
145
Live Deployments
9
Rust Crates
25,911
Total Downloads
14
Years Active
14
Custom Domains
Featured Projects
Our flagship open source projects powering agentic development worldwide.
Explore the project pages — what ships, honest status, install and links →
Turbo Rig
Turbo Rig v5 — the successor to Turbo Flow. A rig that orchestrates AI coding agents instead of replacing them: a cross-family code-review gate where a different model family reviews every diff, git-versioned agent memory, and an AGENTS.md constitution. 8 thin scripts in bash + Python, zero daemons, human-only merges. The gate applies a published finding: in a controlled study of 116 tasks (Xiang et al., arXiv:2607.21656), cross-model review lifted pass rates from 71.6% to 89.7%, outperforming same-family self-review.
turbo-flow
The rules layer for AI-written code. Versions 1-4 were an agentic development environment (Claude Code + Ruflo, 215+ MCP tools, 60+ agents at the v4 peak); v5 is rig-lite — a portable governance kit (cross-model review gate, constitution, git-versioned memory). By Adventure Wave Labs.
Interactive Rust project generator with intelligent templates and enterprise features. 7 project types including CLI, API server, WebAssembly, game engine, embedded systems, workspaces.
Regex-based scanner for AWS keys, GitHub tokens, API secrets. 51k files/sec throughput, 99% detection accuracy, obfuscation detection, enterprise-grade security scanning.
Matrix-style network packet monitor with real-time threat detection. 212x faster packet parsing, DDoS/port scan detection, stunning terminal visualizations built in Rust.
Rust-based agentic loop generator for building autonomous AI workflow pipelines.
Intelligent security sentinel agent for continuous threat monitoring and automated incident response.
A showcase of agentic DevOps workflows combining AI-driven automation with modern platform engineering practices.
AI-powered self-healing infrastructure agent using Ansible for automated remediation and configuration drift correction.
Intelligent code analysis and agentic code review platform for automated quality gates and security scanning.
Automated preflight checks and integration testing framework for agentic deployment pipelines.
Lab Capabilities
Deep expertise across the full stack of modern engineering — from low-level Rust systems to high-level AI orchestration.
Agentic AI Systems
Multi-agent orchestration, autonomous workflows, and AI-powered automation with Claude, GPT, and custom models.
Rust & Systems
High-performance CLI tools, cryptographic libraries, and security scanners built in memory-safe Rust.
Platform Engineering
Golden path developer onboarding, GitOps workflows, and infrastructure automation with ArgoCD & Terraform.
DevSecOps
Security scanning, credential detection, and zero-trust infrastructure with enterprise-grade tooling.
Cloud & DevOps
Multi-cloud deployments on AWS, GCP, and Vercel with Docker, Kubernetes, and Ansible automation.
Developer Experience
Interactive project scaffolding, CLAUDE.md configuration generation, and developer tooling.
Drop an HTML file, ship a live site
A zero-config static deployment pipeline. Drop a standalone HTML file into the sites/ directory of the html-sites repo and CI/CD takes it from commit to a live Vercel deployment — then auto-publishes it to the Turbo Launchpad directory. No build steps, no manual deploys.
Language Distribution
The languages powering our projects.
GitHub Repositories
All open source work from both labs.
Showing 19 of 19 repositories
The rules layer for AI-written code. Versions 1-4 were an agentic development environment (Claude Code + Ruflo, 215+ MCP tools, 60+ agents at the v4 peak); v5 is rig-lite — a portable governance kit (cross-model review gate, constitution, git-versioned memory). By Adventure Wave Labs.
AI-powered developer onboarding platform using natural language to deploy applications via GitOps. Golden path workflows with ArgoCD and OpenRouter integration.
Interactive setup wizard that generates project-specific CLAUDE.md configuration files for Claude Code integration. Supports 12 app types, 7 development methodologies, and 19 feature sets.
Interactive Rust project generator with intelligent templates and enterprise features. 7 project types including CLI, API server, WebAssembly, game engine, embedded systems, workspaces.
Regex-based scanner for AWS keys, GitHub tokens, API secrets. 51k files/sec throughput, 99% detection accuracy, obfuscation detection, enterprise-grade security scanning.
Matrix-style network packet monitor with real-time threat detection. 212x faster packet parsing, DDoS/port scan detection, stunning terminal visualizations built in Rust.
To build an AI agent that accelerates Kubernetes platform development. A platform engineer will provide a high-level, natural language request to create a new Kubernetes API.
Rust-based agentic loop generator for building autonomous AI workflow pipelines.
Intelligent security sentinel agent for continuous threat monitoring and automated incident response.
A showcase of agentic DevOps workflows combining AI-driven automation with modern platform engineering practices.
Full-scale agentic platform engineering demo featuring multi-agent orchestration, GitOps, and autonomous infrastructure management.
Agentic SRE pipeline combining AIOps and human-AI collaboration for automated site reliability engineering workflows.
Progressive delivery demo using GitOps principles with canary deployments, feature flags, and automated rollback.
AI-powered self-healing infrastructure agent using Ansible for automated remediation and configuration drift correction.
Intelligent code analysis and agentic code review platform for automated quality gates and security scanning.
Automated preflight checks and integration testing framework for agentic deployment pipelines.
AI-powered order management assistant for automated order processing and customer service workflows in Spanish.
Intelligent store assistant agent providing personalized customer support and product recommendations in Spanish.
Commercial sales assistant agent for lead qualification, proposal generation, and follow-up automation in Spanish.
Live Deployments
145 Production and non-production projects deployed via Vercel — showcasing what is possible to build and deploy using fully autonomous agentic AI agents and workflows.
Showing 7 of 145 deployments
Open Source Rust Crates
Production-ready Rust packages published on crates.io.
Matrix-style network packet monitor with IP tracking, threat detection and real-time visualization
An interactive Rust project generator with templates and common features
Zero-config cryptographic operations for Rust projects with HIVE MIND collective intelligence
A minimal Kubernetes network inspection tool for diagnosing CNI and pod connectivity
A fast and efficient CLI tool for calculating SHA256, SHA1, and MD5 hashes of files with progress indication and colored output
A blazing fast drop-in replacement for fnv hash algorithm with batch processing optimizations
Common Questions
Answers about the labs, Turbo Flow, and how everything fits together.