$ cat ~/notes/readme.md

# whoami

01

Agentic AI

Multi-agent systems that plan, reason, and act on tools.

02

MCP & Tool Use

Composable tool layers and context sharing via the Model Context Protocol.

03

RAG & Memory

Grounded retrieval and structured long-term knowledge for stateful agents.

04

Eval & Guardrails

Production observability, policy checks, and human-in-the-loop review.

I'm an AI Engineer working on enterprise AI agents, RAG, and MCP tooling. I spend most of my time understanding how agentic systems behave in the real world and what it takes to make them reliable.

Lately I've been exploring agent sandboxing and what it really means to give an agent autonomy without losing control. There's a lot of interesting work happening in this space and I'm still figuring most of it out.

I write to think out loud. Notes on things I'm reading, patterns I'm exploring, papers that made me think twice, and mistakes I've made along the way. Not trying to teach anyone, just sharing the notebook and hoping something lands.

Always happy to connect with people working on similar problems.

$ ps --currently

  • Building

    Multi-agent systems with the Claude Agent SDK and custom MCP servers.

  • Researching

    Long-horizon memory and evaluation patterns for stateful agents.

  • Writing

    Field notes on shipping agentic AI inside enterprise environments.

$ ssh --connect