Drew Breunig

Why is Claude an Electron App?

How System Prompts Define Agent Behavior

The Potential of RLMs

The Rise of Spec Driven Development

A Software Library with No Code

2025 in Review: Jagged Intelligence Becomes a Fault Line

Why I Write (And You Should Too!)

How Model Use Has Changed in 2025

Enterprise Agents Have a Reliability Problem

Don’t Fight the Weights

Glimpses of the Future: Speed & Swarms

Enterprise AI Looks Bleak, But Employee AI Looks Bright

AI Companies School Like Fish

Can Chatbots Accomodate Advertising?

Building Castles in the Air, but With Surprise Physics

Making Sense of AI Job Titles

Bottleneck or Bisect: AI-Assisted Coding Will Change Product Management

Initial Thoughts on GPT-OSS

FLUX.1-Krea & the Rise of Opinionated Models

Does the Bitter Lesson Have Limits?

How Kimi K2 RL’ed Qualitative Data to Write Better

How Kimi K2 Became One of the Best Tool-Using Models

How Rewriting Training Data Improved Kimi K2’s Performance

A Comedy Writer on How AI Changes Her Field

Why “Context Engineering” Matters

Delegation is the AI Metric that Matters

AI Creates the Problems it Solves

The Prompt Foreman

Cat Facts Cause Context Confusion

How to Fix Your Context

Prompts vs. Context