Blog

Field notes on software architecture, AWS, AI architecture, and building with Claude.

• Why agents fail at scale — and how we can stop it

  Loops that never end, context that balloons, costs that 1000×, agents with their hands on the controls — the failures that actually bite in production, and the guardrails that tame them.

• Designing agentic systems that don't fall over in production

  A practical framework for putting LLM agents on the critical path: treat the model as an unreliable service, validate every output, make each step observable, bound the blast radius, and cap the loop.

• A cost-aware reference architecture for serverless on AWS

  How I structure Lambda, EventBridge, and DynamoDB so the bill stays predictable as traffic grows.

• Prompt architecture: treating context like a system you design

  Context windows, tool use, and evaluation loops — the patterns I reach for when building on Claude.

• The seams that matter: drawing service boundaries that hold up

  A field guide to decomposing a monolith without trading one kind of pain for a worse one.

• Evaluating LLM features without fooling yourself

  Building eval harnesses that catch regressions before your users do — and what to measure.

• Multi-account AWS without the operational headache

  Org structure, guardrails, and a landing zone setup that scales with your teams.

• Designing for change: the only constant in real systems

  Why I optimize for the second version of every decision, and how that shapes the first.