Building agentic developer systems, OpenClaw tooling, production SaaS products, and AI-powered automation workflows.

Full-stack technical founder based in Brisbane, Australia. I build production SaaS products, AI automation systems, and open-source developer tools. My current focus is turning AI coding agents from "pair programmers" into reliable engineering systems: structured tasks, atomic commits, review packs, CI loops, and production-safe workflows. I'm currently operating across SaaS, open-source, and agentic development infrastructure — shipping with Codex, OpenClaw, Claude, OpenAI, and disciplined automation.

• 🐎 Thoroughbreds.ai — Equine breeding intelligence platform Next.js · Python · ML — CTO & Co-Founder
• 🧰 CrewCMD — (Opensource) OpenClaw management layer for orchestrating agents, repos, PRs, and review workflows
• 📮 PostDropr — Lead generation from creative content Next.js · Supabase · Stripe
• 🧠 Soply.ai — Knowledge management for growing teams Next.js · TypeScript · AI
• 🎥 ClutchCut — AI product video generation and analysis pipeline Next.js · FastAPI · Turborepo

I build, ship, and operate products end-to-end — increasingly with agentic workflows that turn ideas into tested, reviewable, production-ready changes.

Frontend      Next.js · TypeScript · React · Tailwind CSS
Backend       Python · FastAPI · Node.js · Supabase · PostgreSQL
AI/ML         OpenAI · Claude · LangChain · Vector DBs · NLP · Recommendation Systems
Agents        Codex · OpenClaw · Claude Code · AGENTS.md · Review Packs · CI Loops
Payments      Stripe · Better Auth
Infra         Vercel · Cloudflare · Turborepo · Railway · GitHub Actions
Workflow      Atomic Commits · Branch-Based PRs · Agent Task Queues · Production-Safe Automation

I use AI agents as an engineering force multiplier, not as a replacement for engineering judgment.

Current workflow principles:

Structured task briefs
→ isolated branches
→ atomic commits
→ focused PRs
→ automated verification
→ review packs
→ human approval for risky changes

I'm actively building an agentic development stack where the pieces fit together:

• Agentic Team Playbook: shared operating model, policies, and workflow templates for durable AI agent teams
• CrewCMD: OpenClaw management layer for tasks, agents, repos, PRs, review workflows, and org memory
• taskbrief: turns messy ideas, notes, and TODO dumps into structured agent-ready task queues
• branchbrief: turns completed branches into reviewable, auditable briefs for humans and AI reviewers
• repoctx: repository context tooling for agentic development workflows
• repoforge: deterministic repository generation for agent-friendly source repos
• StackForge: agent-friendly scaffolding CLI for apps, APIs, CLIs, and OSS packages
• agentic-oss-template: open-source repo template with maintainer policy, CI, security docs, and agent workflow rules
• oss-ideas: public idea bank and PRD pipeline for the 70-day OSS sprint

Together, these projects cover the loop from idea capture → task shaping → repo setup → agent execution → branch review → production-safe release.

I'm working through a 70-day public OSS sprint focused on agentic software development.

The goal:

• build practical open-source tools
• improve Codex/OpenClaw workflows
• create reusable agent instruction systems
• ship small, useful developer utilities
• document what actually works when using AI agents in production/community repos

The theme: moving from using agents to operating an agent factory.

"Ship beats perfect — but shipped, tested, and reviewable beats chaos."

I build products to solve real problems, put them in front of real users, and iterate quickly. My current obsession is designing systems where AI agents can move fast without destroying auditability, trust, or production safety.

📍 Brisbane, Australia · 🧑‍💻 GitHub: rogerchappel · 🎙 Creator handle: roger-creator · 📧 [email protected]