Chapter 11 · 5 lessons · 561 min
Using AI to Do Real Work
From chat to workflows that actually save hours.
Overview
Where most people stop at 'I asked ChatGPT and it was okay.' This chapter shows how serious operators use AI to compress hours of work into minutes — for writing, research, code, email, meetings, and creative production.
What you'll learn
- Seven concrete workflows that save you 5–10 hours a week, starting today.
- How to feed AI your own voice so output sounds like *you*, not generic LLM mush.
- When to draft with AI vs. when to use AI as an editor.
- How to build a personal prompt library that compounds in value over time.
Who this chapter is for
Knowledge workers, marketers, founders, students. Anyone who writes, researches, plans, or creates for a living and wants a 10x leverage tool without losing their voice.
Lessons in this chapter
All lessons are free and open — start anywhere, in any order.
- 1AI Workflows That Save HoursConcrete recipes for content, code, research, ops — copy and adapt.115 min
- 2AI for Creative & Marketing WorkBrainstorm, draft, generate visuals, ship faster — without losing your voice.113 min
- 3Common Mistakes & How to Avoid ThemThe 5 traps beginners fall into when working on using ai to do real work — and the fix for each.111 min
- 4A Real-World WalkthroughA start-to-finish example of applying using ai to do real work to a real project you can copy.111 min
- 5Mental Models That Actually HelpThe 3 mental models we keep coming back to when things get hard in using ai to do real work.111 min
Glossary
Key terms used throughout this chapter.
- Workflow
- A repeatable sequence of prompts that takes input X and produces output Y reliably.
- System prompt
- Persistent instructions that set the AI's role and rules for a whole conversation.
- Few-shot
- Showing 2-3 examples in the prompt to teach the model the pattern you want.
- Chain-of-thought
- Asking the model to think step-by-step before answering. Improves reasoning.
- Prompt library
- Your saved collection of proven prompts, organized by use case.
Try it yourself
Knowledge sticks when you use it. Do at least one of these before moving on.
- 01.Take your last 5 LinkedIn or Twitter posts. Feed them to an LLM. Ask it to write 3 new posts in the same voice.
- 02.Pick a recurring task you do weekly (status reports, meeting notes, content ideas). Build it into a single reusable prompt.
- 03.Record a meeting (with permission), transcribe with Whisper, summarize with an LLM. Time the savings.
Further reading & sources
Primary, official sources we cross-checked against while writing this chapter.