Posts
The Inheritance Paradox: Tool Scoping in Multi-Agent Systems
The Inheritance Paradox: Tool Scoping in Multi-Agent Systems
This is Part 2 of a three-part series. See Part 0 for the plugin overview and installation, or Part 1 for the development journey.
Abstract: This article presents empirical observations from developing a Claude Code plugin, revealing fundamental patterns in multi-agent AI system design. Through documented development sessions, we identify three key findings: (1) implicit tool inheritance models consistently fail in practice, requiring explicit scope contracts; (2) platform-enforced topology constraints drive the emergence of specific architectural patterns; and (3) agent naming conventions become functional requirements in AI systems. These findings have implications for the design of agent frameworks beyond the specific platform studied…
The Hybrid Architecture Pattern: Lessons from agent-team-creator
The Hybrid Architecture Pattern: Lessons from agent-team-creator
This is Part 1 of a three-part series. Start with Part 0: Agent Team Creator to see what the plugin does, or continue here for the development journey.
I thought building a Claude Code plugin would take a weekend. Three days into wrestling with MCP tool access bugs, three-location file synchronization, and documentation that promised features that didn’t work, I realized I was building something more valuable than a plugin—I was mapping uncharted territory.
This is the story of building agent-team-creator, a Claude Code plugin that analyzes codebases and generates project-specific AI agents. Along the way, I discovered a pattern that turned platform limitations into architectural clarity: the Hybrid Architecture Pattern…
Claude Code Plugin Journey Part 0: Agent Team Creator
Claude Code Plugin Journey Part 0: Agent Team Creator
What if you could point an AI at your project and have it generate specialists who actually understand your stack? Not generic assistants—agents that know your architecture, your patterns, your conventions.
That’s what agent-team-creator does. Let me show you the workflow…
Ml patterns design
Machine Learning Patterns
Hello! First and foremost, I am embarking on a journey to highlight the main aspects presented in the book “Machine Learning Design Patterns” This series is intended for machine learning engineers, data engineers, and data scientists. I will skip the explanation of basic terminology and focus on the nuances of facing the most common challenges when deploying machine learning models. The concept of patterns was introduced in the field of civil engineering/architecture by Christopher Alexander and five co-authors in the book titled “A Pattern Language” (1977). The main idea is that in a situation that happens recurrently, the pattern describes the core solution for this context and can be applied several times for this or similar ones. Here, we will explore the most general patterns that can be applied to machine learning system design…