Lambda in a VPC Is Fine
AWS re-engineered Lambda VPC networking in 2019 with Hyperplane ENIs, eliminating the cold start penalty entirely. The decision to use a VPC should be architectural, not performance-driven.
A Serverless Knowledge portal for Digital Native Architectures (DNA)
AWS re-engineered Lambda VPC networking in 2019 with Hyperplane ENIs, eliminating the cold start penalty entirely. The decision to use a VPC should be architectural, not performance-driven.
Every tool call in an agentic loop replays the full conversation history. Learn why context grows quadratically, how three common tasks compare in real token cost, what extended thinking does to your context window, and how to test a harness's context transparency.
Most developers treat Claude Code as stateless. The gap between mediocre and excellent AI coding sessions is almost entirely context quality. Learn how to compose CLAUDE.md files, the Powertools MCP server, and the AWS Serverless plugin into a rich, layered context system.
The AI agent part was easy. The hard part was shipping a Node.js process inside a native desktop app as a single binary with no dependencies. Node SEA, Tauri sidecars, and a JSON-RPC bridge on stdio.
A hands-on tutorial building a coding agent CLI with the AWS Strands SDK. Learn how the agent loop works, how to configure Bedrock with extended thinking, manage context, and wire up tools. All in ~285 lines of TypeScript.
Claude Code's documentation tells you to use shell environment variables for Bedrock config. This breaks the layered settings system. Learn how to configure model aliases, permissions, and env vars properly using settings files at user, project, and local scope.
Skills teach how. MCP tools teach what. But nobody standardised when an agent should act. Agency capabilities add triggers and suppression to give autonomous agents the judgement they need.
Kiro's new open weight models run at 0.05x to 0.25x the cost of larger frontier models. Learn when to use DeepSeek v3.2, MiniMax 2.1, and Qwen3 Coder Next to optimise your AI coding budget.
Lambda Durable Functions store your checkpoint data safely and securely, but do you really want all your system operators to see credit card numbers?
Using the Kiro IDE on Windows with Windows Subsystem for Linux (WSL) feels like a dark art. If you are like me, you want all your development tools to be in the WSL, and you want the Kiro agents to be communicating with your local Linux instance as well. This article covers everything you need to know to integrate Kiro with WSL on Windows.
That URL looks public, so surely that traffic is going across the public internet? The assumption is completely understandable. The answer is no.
Dive deep into how Web Sockets really work and understand what they actually provide as a part of your architecture. We will talk about common myths and misunderstandings about these useful bi-directional communication channels and cover real world design and development considerations for success.
We all just copy MCP configs from READMEs and hope for the best. run-mcp lets you keep that simplicity while running servers in complete isolation. One word change, full container security.
Kick off those long-running tasks quickly and easily using an Event Driven solution leveraging AWS Web Sockets. No mess, no fuss and there is an entire CDK Construct for you to leverage so you can build your own in minutes - Just add long-running tasks!
A Web Socket primer! Everything you need to know about API Gateway Web Sockets to get started building!
URL Shortener projects feel a little like the “Hello World” of programming. It’s a simple use-case with some minor complexity that goes a little deeper than just writing out “Hello World” to your callers.
The data you log is critical to understanding how your system is working, learn more about what you should log!
Logging sounds pretty simple. But it is more than just printing out some messages! How you Log Matters and this visual tells you WHY!
Did you ever wonder how your logs get into Cloudwatch? It may feel like magic, but it's not!
A lot goes on when your Lambda function emits a log entry. Let's dive in and understand lambda logging and why we all struggle!
Logging for Observability is always harder than it needs to be. Let's take a look at why this is the case!
Have you ever wondered how to tell if your Serverless system is going wrong or have you ever seen what happens when Scalability boundaries clash? We have got you covered - we have created a project with exactly this problem and have a performance test setup so you can watch what happens when Serverless starts to fail.
When you are building with AWS Lambda, you need to understand how it works in different situations. This article covers the different execution modes of AWS Lambda.
Observability feels hard, but it doesn't have to be. We just need to make sure we have the Observability Mindset and think upfront about observability as we code. How do we achieve this? Read this article to find out how simple it really can be!
Observability is a mindset. It's a conscious choice to expose data to enable questions to be answered when you need to ask them. Learn what it is to Get the Observability Mindset and why it is important. You will only be better at Serverless for getting there!
Scalability boundaries are the number one enemy of Serverless. Know how to watch out for them and what to do when they go wrong.
Serverless thinking is not just about Lambda! You need to focus on and be Event Driven.
You need to retrain your brain and start to get into the Serverless mindset. First step: Think Atomic!
Before you get started, we need to clear up what Serverless means. There is a lot of different information out there!
What's next for Serverless DNA? We will periodically publish updates to the public roadmap as new feature ideas get closer to reality. Feel free to submit your ideas to us using our social contacts.
Our mission statement, what we stand for and what we aim to achieve.
Often we like to jump in and get building straight away without learning everything we need to know. Don't do that, and here is why.
We love to jump in and get building ASAP after we create an AWS account but, take a pause and make sure the credentials you use for your AWS CLI are secure!
We ask ChatGPT all the hard questions, read on to see what it said.
Rendering Markdown content with react-markdown is straightforward. Customising how the markdown content is rendered is trickier.
We use Next.JS and react-markdown for rendering markdown content. Rendering code blocks that are responsive can be tricky, mainly due to the way HTML and CSS interact. Learn what we discovered building Serverless DNA.
We asked chatgpt "If I wanted to build a solution with AWS Serverless technology. What do I need to learn and where would I go to learn about it?". Click through to learn what it thinks.
General information about Serverless DNA. Covers what the site is about and why we created it.