Writing
On AI, DevOps, engineering leadership, and the things I'm building.
ADRs are not just documentation. When you are building with AI agents, they become the persistent architectural context that keeps 30 concurrent sessions aligned without constant human supervision.
Observability When Your Team is 30 AI Agents
Traditional monitoring assumes human operators. When you have 30 concurrent AI agents working on a shared codebase, the observability challenges are fundamentally different. Here is how Sol approaches the problem.
Sol: Multi-Agent Orchestration That Recovers From the 2 AM Crash
Running multiple AI coding agents in parallel sounds great until sessions crash at 2 AM and nobody notices. Sol is a Go binary that coordinates concurrent agents on a shared repo, giving each an isolated worktree, crash-durable assignments, and a merge queue with quality gates. The result: spec out the work, walk away, and come back to merged code or clear diagnostics.
A Load Testing Strategy That Outlives Your Current Stack
Load testing is not about the tool. It is about building a methodology that catches performance regressions before users do, regardless of what language or framework you are running.
Building Incident Response Workflows That Actually Get Followed
Monitoring tells you something is wrong. What happens next determines whether it gets fixed in minutes or hours. A practical guide to runbooks, escalation paths, and post-mortems that teams actually use.
Lawn Care
My approach to overseeding, weed control, and top dressing for a thick, healthy yard in the mid-Atlantic region.
Welcome to My Blog
The start of something. A place to write about the things I build, the problems I solve, and the lessons that stick.