Skip to content
Aaron Lynn
Go Back

AI Doesn't Know What "Good" Looks Like. You Do

3 min read

Tilt-shift diorama of a handmade miniature office with two desks: on the left a calm figure reviews neat papers beside an organised filing cabinet, on the right a frazzled figure faces scattered papers and empty drawers, with 'CLAUDE.md' written on a small whiteboard on the dividing wall.

Same tools. Different filing systems. Guess which one has CLAUDE.md.

1. Know What “Good” Looks Like Before You Prompt

You can’t direct AI if you don’t know what a good result looks like.

A marketing director who’s written hundreds of campaigns will get great copy from AI. A random person prompting for “marketing emails” will get generic slop. The AI doesn’t know the difference — you do.

The “anyone can do anything with AI” narrative is dangerous. AI doesn’t replace expertise. It amplifies it.

Domain knowledge + AI skills is the winning combination in 2026. One without the other leaves value on the table.

Worried about AI taking your job?

Your advantage is you know what the output is supposed to look like. Be the person who directs the AI, not the person the AI replaces.

2. Build a Memory System for Your AI Harness

Every time you start a new conversation in your AI harness, you lose all context. You’re back to square one explaining who you are, what your business does, and how you like things done.

It’s like hiring a new temp every single day.

Fix this by building a memory system in your project folder:

  1. Create a CLAUDE.md file (or AGENTS.md for Codex/OpenCode) in your project folder with your preferences, context, and how you work.
  2. Add your style guides and templates. Link them from the main file so the AI references them automatically.
  3. Add your SOPs and process documents the same way.
  4. After each session, update the file with anything new the AI should remember next time.

That’s the simple version. You can get more sophisticated over time — look up Claude rules and custom instructions if you want to go deeper.

The result: your prompts get shorter. Output quality goes up. Feedback loops shrink. The AI gets smarter about your work over time.

This is the difference between explaining your business from scratch every morning and having a trained staff member who already knows your preferences, your tone, and your processes.

I walk through setting up these AI memory systems step by step in The Complete Claude, Claude Code & AI for Work Productivity.

3. Courses

I have a range of online courses that teach business people what they need to know about productivity and AI:

1. Next Level Productivity
A practical, straightforward course that teaches you how to achieve elite-level personal productivity in today’s constantly interrupted world.

2. The AI Agents, Automations and Agentic Workflows Guide
This non-technical course shows business people and non-coders how they can build and use AI agents in ChatGPT and Zapier.

3. The Complete Claude, Claude Code & AI for Work Productivity
A comprehensive course taking you from beginner AI concepts like prompt and context engineering, to cutting-edge AI productivity using terminal-based AI tools like Claude Code for non-coding office work. Usable with ChatGPT, Claude, Gemini and other current LLMs.

4. The AI Playbook
This is my longer premium course on how businesses can deploy AI tools and technology across their processes and teams.

5. ChatGPT for Managers
See how AI can solve complex management challenges in less than 30 seconds. Full prompt library and examples included.

That’s it for this week!

— Aaron


Share this post on:

Your Best Work Has a Time of Day

Hero Mode Hours is a short guide to finding the one or two daily windows when your brain is sharpest — and spending them on work that matters.

No spam. Unsubscribe anytime.


Keep reading

  • Brute Force Your Way to Success

    Success rarely comes from one attempt — it comes from reps. Plus: what professionalism looks like in 2026, and how AI turns a two-week annual review into three days.

  • If You're Tracking It But Not Improving, You're Wasting Time

    Tracking sleep, steps or deep-work hours is pointless unless you act on the data. Plus: why scaling past solo needs systems, and how adversarial AIs sharpen output.

  • A Guide to Systems Thinking

    Systems Thinking is the ability to see and think of the world as systems. It lets you look behind the curtain of how life works, and change things to better suit your goals and purpose. Here's how you do it.


Hero Mode Hours

Every productivity system you've tried worked for about a week.

That is not your fault. Rigid systems break because they assume every hour is interchangeable. You do not have a time problem. You have an energy problem.

Hero Mode Hours is a short guide to finding the one or two windows each day when your brain is genuinely sharp — and putting your hardest work there instead of wherever the calendar has a gap.

One framework. One day of observation. No app required.