Safety-Critical Platform Engineering at Scale
I build platform primitives within a platform engineering org whose mandate is to be a force multiplier for service teams. I own the monitoring platform that standardizes infrastructure monitoring across the fleet, and built a workflow orchestration control plane now supporting millions of executions across development teams in Dublin, Seattle, San Jose, and New York. That platform is on a path to General Availability for all of Amazon, enabling engineers to build safety-adherent workflows that meet Amazon reliability standards.
Pre-execution guardrails that run 11+ concurrent safety checks before every automated infrastructure change.
Fleet-wide observability standardized across multiple global locations with 200,000+ monitors.
Spec-driven development powering 7 specialist AI agents from 137+ structured specifications.
Portfolio Research · Stock Grading
A structured investment committee framework for evaluating equities. Score any stock across five weighted dimensions — growth, valuation, financial health, competitive moat, and analyst consensus — and receive a letter grade with a written committee summary.
Open Tool →Garmin Integration · AI Coaching
Connect your Garmin account to visualize training load, weekly mileage, heart rate zone distribution, and long run history. Chat with an AI coach powered by AWS Bedrock — contextualized to your actual training data.
Open Platform →Latest post
Designing Safety Guardrails for Distributed Workflow Orchestration If you automate changes across a large fleet, you will eventually face the same uncomfortable question: what happens if we run this at the wrong time, ag…