Research & Innovation
An expert AI engineer specializing in the Model Context Protocol (MCP) and local LLM inference on Apple Silicon. They demonstrate deep expertise in building advanced developer tools, agentic workflows, and high-performance machine learning pipelines using both TypeScript and Python. Their work emphasizes "governance-as-code" and bridging the gap between local development environments and sophisticated AI models.
Score reflects a highly capable engineer pushing the boundaries of local AI and agent protocols. While technical innovation is elite (10/10), some lower scores in testing reflect the experimental nature of cutting-edge research work rather than a lack of skill.
The profile presents a very strong, consistent signal of an expert working at the forefront of AI engineering. The repositories are original, substantial, and directly demonstrate the skills analyzed.
Prioritizes exploration over polish
No significant red flags detected.
Fast parallel LLM inference for MLX
Letting Claude Code develop his own MCP tools :)
Claude Deep Research config for Claude Code.
An MCP Server that's also an MCP Client. Useful for letting Claude develop and test MCPs without needing to reset the application.
Agent Engineering course files