Ten prompting patterns I actually use with coding agents: adversarial reviews, blunt boundaries, measurable gates, evidence, project-specific context, and L2 meta-prompts.
Coding agents got much better.
Our prompts often didn't or we use inefficient chat messages.
In 2026, I don't try to find magical sentences that somehow make the model smarter.
I try to make the task hard to misunderstand, hard to game, and easy to verify.
After years of using LLMs on real codebases, these are the ten prompting patterns I keep coming back to.
Part 1 of Practical AI Engineering: Beyond the Demo Most developers think better prompting means...
Most production AI features are still glueing prompts together with string concatenation in some random Node service. The prompt…
Short take: AI coding agents get useful when they have clean context, not giant prompts.
How to move from trial-and-error prompting to a systematic approach for integrating AI into your codebase.
You used to craft the perfect prompt. Tweak the wording. Add examples. Get a better answer. That era...
Moving from simple prompt-response loops to iterative, self-correcting agentic patterns in your LLM applications.
