Hello World

I've been building software for over eight years — at HP, AWS, and now Amazon StoreGen — and I've learned more from the messy parts than the clean ones. The failed deploys, the architectural bets that didn't pan out, the late-night debugging sessions that revealed something fundamental about how a system actually worked.

This blog is where I write about all of that.

What I'll cover

AI agents and multi-agent systems. I'm currently building an autonomous AI agent at Amazon that writes code, joins design discussions, and generates artifacts across the SDLC. I'll write about the architecture behind it — MCP, A2A protocols, tool use, orchestration patterns, and the hard problem of making agents reliable in production.

Distributed systems. I spent four years at AWS designing event-driven smart grid systems — real-time pipelines processing millions of meter events with Lambda, Kinesis, and DynamoDB. I hold a US patent (#12,541,394) for a distributed execution framework that came out of that work. Expect posts on event sourcing, partitioning strategies, consistency tradeoffs, and the gap between distributed systems theory and practice.

Developer tools and infrastructure. I build CLI tools, MCP servers, and automation that makes engineering teams faster. I'll share what works, what doesn't, and the design decisions behind the tools I ship.

Engineering tradeoffs. The interesting part of software engineering is rarely the code — it's the decisions. Build vs. buy. Consistency vs. availability. Simple now vs. flexible later. I'll write about how I think through these choices.

Why now

I've been meaning to write for years. I finally built this site (notebook theme and all), and I'm committing to publishing regularly. Some posts will be deep technical dives, some will be short observations, and some will be walkthroughs of things I built and why.

If any of this sounds interesting, stick around.