Index

Writing

Notes on agentic systems, MCP, developer experience, and AI-assisted tooling. Read in series order where a set of posts belongs together.

06 entriesUpdated continuously
  • Building an AI-Powered CLI for File Operations cover image
    11m read

    Building an AI-Powered CLI for File Operations

    How I designed an AI-powered CLI for file operations using MCP, agentic workflows, and developer-focused automation across local and remote systems.

    AIAgentic SystemsDX
    Read article
  • Surveying the Agent Ecosystem: MCP, LLMs, and Cloud Files cover image
    16m read

    Surveying the Agent Ecosystem: MCP, LLMs, and Cloud Files

    A survey of MCP agents, LLM tooling, and cloud file workflows to identify gaps in real-world file automation across local and remote systems.

    AIAgentic SystemsMCP
    Read article
  • Technologies for a Local-First MCP Host Application cover image
    11m read

    Technologies for a Local-First MCP Host Application

    Why I chose TypeScript, Bun, Docker, Ollama, and a local-first architecture for an MCP host application for AI-assisted file workflows.

    AIAgentic SystemsDX
    Read article
  • Uplink: A Modular MCP-Powered CLI for File Operations cover image
    18m read

    Uplink: A Modular MCP-Powered CLI for File Operations

    How I built a modular MCP-powered CLI for file operations with local and remote agents, tool orchestration, guardrails, and extensible architecture.

    AIAgentic SystemsDX
    Read article
  • Evaluating an MCP Agent: Tools, Latency, and DX cover image
    15m read

    Evaluating an MCP Agent: Tools, Latency, and DX

    A practical evaluation of an MCP agent for file workflows, including latency, tool-scaling costs, workflow comparisons, and developer feedback.

    AIAgentic SystemsMCP
    Read article
  • MCP Agents for File Workflows: Limits and What Comes Next cover image
    10m read

    MCP Agents for File Workflows: Limits and What Comes Next

    An evaluation of MCP agents for file workflows, covering usability limits, model speed, tool scaling, and what still blocks day-to-day adoption.

    AIAgentic SystemsMCP
    Read article