Projects
R&D explorations of what product work becomes when AI handles execution. Each project feeds into The Boundary, where I share what I learn.
Production Projects
Vockify
The complete product operating model: from customer feedback to strategic roadmap, powered by AI
Intelligence
Daily — Auto-organize feedback, daily digests, AI sentiment analysis
Strategy
Quarterly — Map to customer jobs, score opportunities, build solution trees
Execution
Now/Next/Later roadmap connected to opportunities, public sharing, impact measurement
Target: Product Managers at B2B SaaS companies (10-200 employees) making daily prioritization decisions without dedicated customer insights teams.
Active Projects
HatchClaw
Multi-agent system for building software
What happens when AI agents don't just assist product work, but execute entire workflows? A "product team" of AI agents—AI CMO (positioning), AI CTO (technical decisions), AI PMs (discovery and specs), AI engineers (shipping code).
Current focus: Foundation Stage
Building the coordination architecture. The hard part isn't individual agent capability—it's orchestration. Getting multiple agents to work together without chaos is a system design problem, not a prompt engineering problem.
The Boundary
Essays on AI-native product work
Not "AI tool reviews"—documenting what actually happens when you try to dissolve the boundary between product and engineering.
Read essaysConcierge of AI
Managed AI concierge service for high-performers
We deploy, run, optimize, and evolve your personal AI operations layer. You pay for the outcome. We handle everything else. Built on top of HatchClaw.
Key insight
High-performers don't want to manage AI tools—they want results. The value is in operating AI for them, not giving access to it.
How these connect
Each project tests a piece of the same broader question: What does product work look like when AI is the executor?
HatchClaw
Multi-agent execution
The Boundary
Public learning
Concierge
Service delivery
Vockify
Workflow automation
The insights feed into each other. What I learn about agent coordination in HatchClaw informs how I think about AI-assisted research. The questions readers ask about The Boundary shape what I test next.
Building in public
I share what I learn, including failures. But I'm thoughtful about boundaries:
Metrics shared
User numbers and learnings
Privacy protected
All user learnings anonymized
Strategic boundaries
General strategies shared, tactics private
The best way to track these projects:
Subscribe to The Boundary