⚡ Level 3 of 4

AI Practitioner

"AI is part of my daily workflow — I'd miss it if it were gone."

What this level looks like

You use AI daily, often in multiple ways. You have a preferred AI assistant and may also use productivity tools, image generation, or voice/dictation. You treat the AI as a dynamic thinking partner rather than just an answer machine. You are beginning to master precise prompting, apply rigorous critical thinking to verify outputs, and identify exactly where AI enhances—or hinders—your specific workflow. But you might still be falling into the trap of writing huge prompts or manually copying prompt templates. To cross the threshold to advanced, move away from manual prompting towards customizing your tools with persistent context.

What to do next

✓ Move away from manual copy-pasting of long context or templates. Set up a dedicated Claude Project or Custom GPT with pre-loaded context files (product descriptions, style guides)
✓ Practice the Human-AI Sandwich: frame the problem yourself first, let AI generate as a thinking partner, and then rigorously evaluate, edit, and fact-check the output
✓ Explore reasoning models with extended thinking/effort settings enabled when you need help with complex analysis, strategy, or coding

Recommended guides

The Anthropic Ecosystem: A Complete Guide to Claude's Products and Models

Everything you need to know about Anthropic's Claude — from the model family (Haiku, Sonnet, Opus) to the web app, connectors, Claude Code, and Claude Cowork. A practical guide from someone who uses these tools daily.

15 min read

Let the AI Interview You: How to Extract What You Actually Know

The biggest bottleneck in making AI useful isn't technology — it's getting your knowledge out of your head and into a format AI can work with. Here's a method that works, from a five-minute technique to a full team knowledge extraction process.

13 min read

The Research Pipeline: How I Use Perplexity and Claude Together

Most people use AI tools in isolation. Here's the workflow I use daily — Perplexity for sourced research, Claude for deep analysis — and why the combination is worth more than either tool alone.

13 min read

Your AI Agrees With You Too Much

AI assistants have a dangerous habit: they validate whatever you say. That's not a bug you can fix with 'be more critical.' Here's what actually works — and the three failed attempts that taught us why.

15 min read