Recent content in sjramblings.io — Stephen Jones on AWS, AI Agents & Cloud Architecture on sjramblings.io
If you’ve ever opened a pull request to add one more Service Control Policy to your Landing Zone and watched it fail with “Max Allowed SCPs for ou ‘Infrastructure’ is 5, found total 6”, you’ve felt the pressure I’m about to talk about.
I had 47 tabs open. AWS What’s New, a couple of company blogs, three podcast feeds, two YouTube subscriptions, X bookmarks, Substacks I’d been meaning to read. None of them were going to read themselves. That’s the moment I knew I had a problem the existing tools weren’t solving.
AWS publishes a JSON file every day that nobody reads. It’s at ip-ranges.json . 2.4MB, no story.
An Australian lawyer asked me last week whether he could run “frontier Claude” on AWS Bedrock in Sydney and keep his client data in Australia. Simple enough question. The answer is one of those ones where you open your mouth ready to say “yes, obviously” and then close it again, because as of April 2026 the honest answer is “not the way you think, and here is the bit…
In March 2026, AWS Network Firewall became available in the European Sovereign Cloud. If you’re not paying close attention to the sovereign cloud space, that probably sounds like a minor service addition. I’d argue it’s one of the most significant milestones since the partition launched.
This is Part 3 of a three-part series on building skills for Claude. Part 1 covered what skills are and why they matter. Part 2 walked through building and testing your first skill. Now we get into the stuff that separates a toy skill from a production one: proven patterns, distribution mechanics, and how to fix things when they break.
I asked my agent to find a deployment guide. It loaded 47 documents, burned 23,000 tokens, and returned the wrong one.
Welcome back to the series. In Part 1 , we covered what Claude Code skills are, why they matter, and how they transform Claude from a general-purpose assistant into a specialist that knows your workflows. Now it’s time to build one.
This is Part 1 of a 3-part series on building skills for Claude. If you’ve ever wished Claude could just remember how you like things done, this series is for you.
A few weeks ago AWS dropped Agent Plugins , a packaging model that bundles skills, MCP servers, hooks, and reference docs into installable units for AI coding agents. Two commands and your Claude Code or Cursor agent knows how to deploy to AWS, estimate costs, and generate IaC.
AWS quietly dropped a prescriptive guidance document on MCP strategies this month. If you have spent any time with MCP servers, you know the protocol itself is straightforward. The hard part is everything around it: how many tools to expose, where to host the servers, how to stop an agent from deleting your production database with inherited admin credentials.
Most people building with AI agents hit the same wall eventually. Your agent forgets things between sessions. RAG retrieval surfaces the wrong chunks. You cannot tell why the agent picked what it picked. And every time you start a new session, you are re-explaining context that should already exist.
Your team just got the green light to build an AI agent for customer support escalation. The architect says “CDK and AgentCore.” The PM says “show me something by Friday.”
Every time I set up log shipping from a non-AWS source to CloudWatch, the same friction shows up. Install an agent. Configure IAM credentials. Implement SigV4 signing. Manage rotation. It works, but it is a lot of ceremony for “send this text to that place.”
We’ve spent the last two years watching AI coding assistants evolve from glorified autocomplete to genuine collaborators. But there’s been a persistent gap between “AI that helps you code” and “AI that ships features while you sleep.” On March 5, 2026, OpenAI quietly open-sourced the bridge: OpenAI Symphony .
Last week, Chris Porter and I entered the Airia F1 Atlassian Williams Racing virtual hackathon as team JoPoCo (Jones + Porter + Co). We had seven days to build an AI-powered fan experience for remote motorsport viewers. What we shipped was Race Radio Control: a second-screen companion that lets fans tune into four AI-powered radio feeds — Carlos Sainz, Alex Albon, team principal James Vowles, and…
Another week, another quantum breakthrough. On March 2, the Advanced Quantum Technologies Institute announced the JVG algorithm, a hybrid classical-quantum approach that, they claim, could factor RSA keys with fewer than 5,000 qubits. That is three orders of magnitude below the million-qubit estimates we have been working with for years.
AWS quietly announced support for 30 new resource types in AWS Config on March 2, 2026. If you’re the kind of person who skims these announcements and moves on, I get it. Most Config resource type expansions are incremental. This one isn’t.
What if everything we know about RAG is built on a flawed assumption?
On 18 February 2026, AWS quietly updated ACM to reduce the default validity of public certificates from 395 days to 198 days. If you’re running anything on AWS that terminates TLS — CloudFront distributions, Application Load Balancers, API Gateway endpoints, Elastic Beanstalk — this affects you. If you’re managing secrets alongside certificates, your Vault deployment should be part of…
I’ve been staring at a Terraform module for the last ten minutes, and I can’t stop thinking about a question that would have been absurd two years ago: why am I writing this?
If you’ve been building AI workloads in Australia, you’ve felt the frustration. The models you want to use are sitting in US regions. Your compliance team is asking where inference data is being processed. And every API call is adding 180-200ms of network latency before the model even starts thinking. Run a five-step agentic workflow and you’re adding a full second of pure…
If you’ve ever needed to run a hypervisor inside an EC2 instance, you know the pain. For years, the answer from AWS was simple: buy a bare metal instance. That meant paying for an i3.metal or m5.metal just to get access to hardware virtualisation extensions. Need to test a Firecracker microVM setup? Bare metal. Want to run KVM for a security sandbox? Bare metal. Running nested Hyper-V for a…
I spent the last week falling down a rabbit hole. Not the productive kind where you emerge with a working solution and a sense of accomplishment. The kind where you save ten GitHub repos in a single week and then sit back and realise they’re all telling you the same thing.
The Decision You’ve Been Making on Autopilot Every AWS practitioner has a version of this conversation at least once a quarter. Someone on the team suggests building something custom. Someone else points out there’s a managed service or SaaS product that does it already. The room does the mental maths: engineering time, ongoing maintenance, opportunity cost. Nine times out of ten, you…
The Number That Should Worry You AWS raised GPU Capacity Block prices by 15% on a Saturday in January. No blog post. No announcement. Just a quiet update to the pricing page that said prices were “scheduled to be updated” without mentioning which direction.
You’ve Got the Tools. So Why Are You Still Slow? If you’re building on AWS right now, you have access to more managed services, more abstraction layers, and more AI-assisted tooling than at any point in computing history. CDK, SAM, Amplify, Bedrock, Kiro, Claude Code. The list keeps growing.
If you’ve ever written an IAM trust policy for GitHub Actions OIDC federation, you’ve probably done the thing we all did. You set the sub condition to repo:my-org/my-repo:* , told yourself “that’s scoped enough,” and moved on with your day.
My dad had a camera. Not a phone with a camera. A camera. A proper one with a roll of film that gave you 24 shots, maybe 36 if you were feeling extravagant.
Builders are living in a weird new era. An agent can scaffold a project, wire up an API, generate tests, refactor the mess, and ship a feature before you finish your second coffee.
Rob Kennedy, AWS Vice President of Network Services, opened his re:Invent 2025 keynote with a simple metaphor: atoms bond into molecules, molecules combine into structures, and those structures become complex organisms. The implication was clear, networking is no longer about connectivity. It’s about intent .
Many organisations are rushing into deploying AI agents with the same enthusiasm they had for serverless in 2016, great technology, terrible operational discipline. The pattern is predictable: build a proof-of-concept that works brilliantly in a demo, deploy it cautiously to production, then discover you have no idea how to govern what it’s actually doing once users interact with it at…
I watched Werner Vogels deliver what he’s calling his final AWS re:Invent keynote, and it struck me that he wasn’t talking about new services or feature announcements. Instead, he spent an hour articulating why the tools matter less than the person holding them. After 14 years of keynotes, Amazon’s CTO decided to hand the microphone to younger voices—but not before leaving…
If you’ve been managing AWS Transit Gateway costs in a multi-account environment, you’ve probably felt the pain of sender-pay billing. Well, I’ve got great news, AWS just launched Flexible Cost Allocation for Transit Gateway and your FinOps team is going to love this!
If you’ve been building with LLMs lately, you’re probably as excited as I am about the possibilities! But let me tell you about something that’s been keeping security folks up at night… prompt injection vulnerabilities.
Ever wondered what it’s like to use an AI tool to improve itself? I just spent 2 hours using Amazon Q CLI to build a new feature for Amazon Q CLI, and the experience was genuinely mind-bending.
Working with a single AI assistant on complex projects is like having one engineer handle an entire software delivery pipeline. Possible? Sure. Optimal? Not even close.
Summary AWS just flipped the switch on their newest region: Asia Pacific (New Zealand) - ap-southeast-6 . After years of routing traffic through Sydney, Kiwi organizations finally have a local AWS presence. This isn’t just about national pride — it’s about single-digit millisecond latency, data sovereignty, and unlocking cloud-native architectures that were previously cost-prohibitive.
Summary AWS just dropped a networking feature that’s going to change how we think about VPC routing forever. VPC Route Server brings dynamic routing capabilities directly into your VPC, automatically handling failover scenarios that used to require complex scripting or third-party solutions. If you’ve ever wrestled with static routes and manual failover for network appliances, this…
Daniel Miessler just dropped a fascinating deep-dive into building what he calls a “Personal AI Infrastructure” (PAI) - essentially an AI-powered life management system that handles everything from content creation to security assessments. While his approach uses Claude Code and local tooling, it got me thinking about how we could architect something similar using AWS services.
Daniel Miessler just published something that made me stop and think: “What are we actually building with all these AI tools?” It’s a question that cuts through the hype and gets to the heart of what matters.
Let’s be honest. The most exciting part of building an AI agent is the agent itself—the logic, the prompts, the creative problem-solving. The least exciting part? The ceremony. The boilerplate. The tedious dance of wrapping our code in an API, writing a Dockerfile, managing ECR repositories, and wrestling with deployment scripts to get our creation into the cloud.
There’s something magical about the games we grew up with. The simple mechanics, the blocky graphics, and the maddeningly catchy music are etched into our collective memory. So when AWS announced the Build Games Challenge , a global event to recreate these classics using modern AI tools, I knew I had to jump in.
Recently I’ve been trying out n8n as a workflow automation tool and I’m really enjoying the flexibility it offers. Of course, being an AWS Community Builder I would naturally run this on AWS Fargate as the n8n software is available as a container , however to keep the costs down I ended up running it on Amazon Lightsail.
Hey there, tech enthusiasts! Ever stared at your AWS bill and wondered, “Where did that come from?” Yeah, me too. Especially when diving deep into services like fsx for NetApp ONTAP and considering the magic of s3 Intelligent-Tiering to keep those storage costs in check.
Hey there, Tech Friends! 👋 Let’s talk about something that might not sound super exciting at first glance, but trust me, if you’re wrestling with cloud infrastructure, especially in regulated industries, this is pure gold. We’re diving deep into the newly announced Time-based Copy for Amazon Machine Images (AMIs) .
Unsealing Vault after a server restart is one of those tasks that, while essential, can quickly become cumbersome—especially when you’re managing multiple Vault nodes or clusters. Enter Auto Unseal, a lifesaver in production environments, as it automates the process of securely unsealing Vault without requiring manual intervention.
Hey there, tech aficionados! 👋 Recently, AWS dropped a neat minor update – they started showing the full size of your EBS snapshots. Game changer! This isn’t just the incremental stuff; it’s the data footprint.
Hey there, tech enthusiasts! Ever felt that little thrill of power when you get root access on a system? It’s like holding the keys to the kingdom, right?
Today, I was reminded of the rich content stored in AWS Config and how easily it can reveal so much about an AWS Organisation across one or many accounts.