Based in New Zealand · AWS · AI Engineering
Raj Murugan
Solutions Architect at Parallo (Crayon Group). I build production AI agents on AWS — real systems with persistent memory, guardrails, and CI/CD pipelines. Not demos.
Currently writing: a 6-part series on building production AI agents with Amazon Bedrock AgentCore — the real gotchas, CDK stack, Strands SDK, and cost breakdown. Read it →
About
I'm a Solutions Architect at Parallo, a Crayon Group company, where I design and ship cloud-native applications on AWS for enterprise clients.
My current focus is production AI agents — moving past demo chatbots to real systems with persistent memory, guardrails, cost controls, and CI/CD pipelines. I've shipped a full Bedrock AgentCore deployment from zero to production and documented every trap along the way.
This blog exists so you don't have to hit the same walls.
Stack
Featured Work
Production-ready Customer Service AI Agent on Amazon Bedrock AgentCore. Every deployment gotcha documented inline.
Recent Posts
View all →Part 6: Cost & Performance for Bedrock AgentCore — Prompt Caching, Model Selection, and CloudWatch Alarms
Real cost breakdown of running an AgentCore agent: prompt caching savings, when to use Nova Pro vs Claude Sonnet, PriceClass_100, idle timeouts, and how to set alarms before your bill surprises you.
Part 5: CI/CD for Bedrock AgentCore with GitHub Actions and AWS OIDC (No Stored Credentials)
How to build a complete CI/CD pipeline for AgentCore using GitHub Actions OIDC: no stored AWS keys, dual-tag ECR strategy, automated Runtime updates, and multi-environment promotion.
Part 4: Running Your AgentCore Agent Locally with Docker (The Right Way)
How to build and run your AgentCore container locally with real AWS credentials, the correct linux/amd64 platform flag, the .env.local pattern, and how to test with curl.