dennistraub

AI coding agents amplify whatever vocabulary you give them. Domain-Driven Design's Ubiquitous Language is the pattern that makes that amplification work for you, not against you.

Two contradicting takes on AI and junior developers aren't actually contradicting. One describes individual disposition, the other organizational design. The real risk is making pipeline decisions by accident.

Developers are reinventing Bounded Contexts and Anti-Corruption Layers for MCP servers without knowing it. The patterns already exist - and they've been battle-tested for 20 years.

Parallel AI agents write correct code that doesn't integrate. Fix it by extracting shared contracts before generation and running a review agent across all boundaries after.

Parallel AI agents write correct code that doesn't integrate. Fix it by extracting shared contracts before generation and running a review agent across all boundaries after.

MCP vs CLI vs directly calling the API, the real question is the same: where do the credentials live when agents scale?

Three small things that tripped me up when setting up a Python monorepo with uv workspaces - root naming, inter-package deps, and pytest collisions.

The Strands Agents SDK has a built-in persistence layer for conversation history - just pass a SessionManager to the Agent constructor.

From Zero to Agentic Coding: Running...

This is my third major technological shift, and every time I hear the same question echo through the...

An important part of my job is to collect and distill feedback into recommendations for product and...

In this walkthrough, I'll show you the simplest way to deploy an MCP server on AWS, using the Amazon...

This is a full tutorial with detailed explanations. For a quick walkthrough, see the code-only...

Top 7 Featured DEV Posts of the Week ...

When Anthropic published the Model Context Protocol (MCP), I immediately started experimenting with...

Welcome to the third part of our series on building AI agents with Strands! In Part 1: Creating...

Welcome to the second part of our series on building AI agents with Strands! In Part 1, we created a...

In this first tutorial, I'll walk you through creating a simple but functional AI agent using the...

Welcome to this hands-on tutorial series on building AI agents with the Strands Agents...

Another great session at this weeks German OWASP Day was the closing keynote by Christina Lekati with...

I just got back home from the German #OWASP Day, and one of the great presentations was Björn...

No matter if you build your workloads in the cloud or on-premises, there are a number of principles...

Zero Trust is a security model and set of mechanisms that emphasizes identity verification, least...

Sasha Costanza-Chock just tweeted the following in response to the fact that OpenAI, while...

SQL injection attacks are one of the most common and dangerous types of attacks on websites and apps....

Amazon S3 is often used for critical data such as backups and business-critical files. Protecting...

Amazon S3 is known for its reliability, scalability, and low-cost storage options. However, despite...

AWS has just announced the release of two new AWS managed policies for Amazon Detective –...

Input validation is a technique used to ensure that input data is properly formatted and within an...

As a developer, it is essential to ensure the applications you build are secure and protect sensitive...