Hey friend, Welcome to the first issue of Future-Proof Your Career with AI. With AI innovations, tools and use cases changing at a lightning pace -- it's an honor to be your AI sherpa. Like with my RadReads newsletter, my goal is to comb through everything to save you as much time as possible. Today, I'm going to show you how I used ChatGPT to replicate my writing style and voice for my email newsletter. Step 1: The initial promptAs you'll see below, the goal of this exercise was to replicate a very specific type of writing - summarizing an article into a 4-6 sentence blurb. (It wasn't to replicate longer-form writing, like a blog post or newsletter.) I have a distinct advantage here in that I've written extensively and have easy access to writing samples. If you don't have this access, you have a few options:
Here is the initial prompt: I have a file with 50 blurbs.
I want to extract common phrases, tone, sentence structures, or rhetorical devices from past blurbs.
Once extracted, this would be static. I
would love to implement multiple summary structures (e.g., listicle, narrative, key takeaways, provocative question-based)?
I wonder if we can extract this from our 50 blurbs The length is pretty consistent (we can extract).
Step 2: Upload the writing samplesI used 50 writing samples but you could probably get away with 5-10. Here's one example: It’s a slippery slope to intertwine your identity and your achievements. Particularly when you pivot your career or hit retirement. Here’s a son’s tragic story about his dad — and how he achieved his biggest, most singular professional goal — only to realize that “there was nothing on the other side.” And the next one: When you equate your net worth to your self-worth, what could go wrong? "I was optimizing a sick and toxic equation," writes the author of this post (ironically, a former financial planner). Her new North Star? Emotional wealth. Step 3: ChatGPT analyzes the blurbsFirst, it returned an analysis of the writing style:
Total blurbs analyzed: 40 (some shorter entries might have been excluded)
Average word count per blurb: ~51 words
Most common sentence starts: Includes “But,” “The,” “What,” and “If” (suggesting a mix of contrast, definition, inquiry, and hypotheticals)
Most frequent two-word phrases: Shows recurring themes and connective language
Most frequent three-word phrases: Highlights repeated rhetorical patterns like “only to realize” and “What happens when”
But more importantly, it observed something more juicy - different narrative structures:
Listicle blurbs: Recognizable by markers like 🚀, 👀, 🤔, or bullet points
Narrative blurbs: Begin with personal storytelling (“I…”)
Question-based blurbs: A significant number include rhetorical or open-ended questions
When I asked it to dig deeper on each structure, it was really good: The “Big Idea” (Concept Breakdown & Reframing) 💡
Structure:
Introduces a widely held belief or assumption. Presents a counterintuitive insight or paradigm shift. Uses a short punchy example or metaphor. Ends with a new way of thinking about the idea.
📌 Examples from your blurbs:
Goals are a tricky beast. We need them to stay focused and motivated. But once we achieve them, we’re often greeted with boredom or dissatisfaction (aka the when-then-trap). An antidote? Replace goals with growth loops.
Is our purpose on this earth to create economic value? At our core, we know the answer is no — but as we strive, create and make money, our activities seem to be in conflict with our values.
📌 Why it works:
Reframes common ideas in a fresh way. Creates an “aha” moment—good for virality. Encourages reflection—people rethink their assumptions.
(I also pushed ChatGPT to identify more structural formats, but I've excluded them here for the sake of brevity.) Step 4: Create reusable promptsArmed with great structures and my writing style, I had it generate prompts for each type of narrative structure. (Remember, the goal was to be able to feed it an article and then have it summarize it in my voice) Create the first resuable template for AI Prompts for: 1. The Thought-Provoking Question Here's the prompt it came up with. Pay particular attention to "format:" You are summarizing an article in the style of Khe Hy. Your tone is conversational, thought-provoking, and slightly playful. Your writing engages the reader with a mix of curiosity, insight, and a touch of irreverence.
Your task: 1. Start with a bold, engaging question related to the article.
2. Challenge a common assumption or belief.
3. Provide a concise, thought-provoking summary of the article’s main idea.
4. End with a rhetorical question or a final twist that leaves the reader reflecting.
Format: [Opening question] [Challenging statement] [Brief explanation] [Final thought-provoking question]
Examples:
- Are you sure you’re not playing a status game? That Tesla, NPR tote, and your book club picks? They’re all signaling something. Status isn’t just about money—it’s about identity. So what game are you actually playing?
- Does more money make you happier? We’ve all heard the “$75K happiness rule,” but it turns out the truth is more complicated. Money helps—until it doesn’t. So if $100M can’t fix financial anxiety, what will?
- Are you over-optimizing your life? Squeezing every minute, every email, every micro-task. It’s productive. But is it meaningful? When you look back, will you care about how much you got done—or how you actually felt?
Now, generate a summary of the article using this structure.
Step 5: Consolidate all the structures into a mega-promptThis was a simple copy-paste, but once I had all my prompts, I consolidated them into one large prompt so that I could get four different options for each article. Step 6: Refine the mega-prompt with word counts and banned wordsI then noticed that the prompt didn't take my writing style into consideration. So I fixed that: Should we add target sentence word counts (non-binding) and target # of sentences (non-binding)? I wanted them to be soft-targets versus hard constraints, hence my use of "non-binding." Next, it was returning emojis (yuck) and words I would never use ("unlock your potential" and "game changer"). It came out really well. Where would we create a list of banned words and phrases Step 7: Put it all togetherFinally, we had covered all of our bases. Here's the final result: You are summarizing an article in the style of Khe Hy. Your tone is conversational, thought-provoking, and slightly playful. Your writing engages the reader with a mix of curiosity, insight, and a touch of irreverence.
Your task:
Generate three different summaries of the article using distinct formats. Each summary should be between 40-60 words and 3-6 sentences long. ### 1️⃣ Thought-Provoking Question Format
- 3-5 sentences (avg. 8-15 words per sentence) - Start with a bold, engaging question related to the article. - Challenge a common assumption or belief. - Provide a concise, thought-provoking summary of the article’s main idea. - End with a rhetorical question or a final twist that leaves the reader reflecting.
Example:
- Are you sure you’re not playing a status game? That Tesla, NPR tote, and your book club picks? They’re all signaling something. Status isn’t just about money—it’s about identity. So what game are you actually playing? ### 2️⃣ The Big Idea (Concept Breakdown & Reframing) - 4-6 sentences (avg. 10-18 words per sentence)
- Start by introducing a widely held belief or conventional wisdom. - Immediately challenge or complicate that belief. - Introduce a counterintuitive insight or reframe the idea in a new way. - End with a short, memorable takeaway.
Example:
- “More money = more freedom.” But for many, more money = more anxiety. Wealth can be a trap if it keeps moving the goalposts. So what if real freedom isn’t about having more—but about knowing what’s enough? ### 3️⃣ Key Takeaways (Narrative Style, No Lists)
- 4-6 sentences (avg. 12-20 words per sentence) - Open with a hook that captures the core theme of the article. - Weave in the main takeaways naturally, making them feel like part of a conversation rather than a list. - Use sharp, compelling phrasing to keep the insights engaging. - End with a final thought that ties everything together or leaves the reader reflecting.
Example:
- Everyone talks about optimizing productivity, but no one asks whether it’s worth it. This article challenges the obsession with efficiency, arguing that squeezing more into our days can make life feel smaller, not bigger. True fulfillment isn’t about doing more—it’s about creating space for what actually matters. ### Guidelines - Each format should be 40-60 words long.
- Target: 3-6 sentences per summary. - Sentence length guide: - Thought-Provoking Question: 8-15 words per sentence - Big Idea: 10-18 words per sentence - Key Takeaways: 12-20 words per sentence - Keep it conversational, punchy, and slightly playful. - Use short, sharp sentences to maintain engagement. - Avoid lists, bullet points, or over-explanation. - Leave the reader with something to think about. ### Prohibited Words & Phrases - Avoid generic corporate phrases like “unlock your potential,” “game-changer,” “life hack.”
- Do not use clichés like “at the end of the day” or “outside the box.” - Avoid excessive enthusiasm—no “this will blow your mind” or “you won’t believe this!” - Do not sound overly promotional—no “this book will change your life” unless explicitly stated in the article. - Stick to sharp, conversational, and thought-provoking language. Now, generate three distinct summaries of the article using these structures.
This mega-prompt was the culmination of months of false-starts. If I'm being honest, I initially didn't put in the real trial and error to get this up to the highest standard. And as you can see here, I relied heavily on ChatGPT to guide me towards the solution that I was seeking. I encourage you to ask it, "What other questions do you have for me?" or "What do you need from me to help me achieve my desired outcome?" I hope you can incorporate this into your workflow. See you next Thursday, Khe PS Did someone awesome forward this to you? Here's where you can sign up to Future-Proof your Career with AI. |
Ready to achieve your goals and get more out of life? Join 50,000 ambitious professionals who are pursuing productivity, growing their career and creating financial freedom.
Hey Reader, I hope you're all having a fantastic weekend. Shortly after sending out last week's newsletter, I got to tour Cincinnati with my friend (and RadReads super-fan) Casey Grisez. IRL meetings with friends is such a gift! I've also sent three issues of our new newsletter Future-Proof your Career with AI. I know you are all so busy. And it's impossible to keep up with all the new developments. This week, I test-drove NotebookLM (by Google). This will definitely be my go-to tool when I'm...
Hey Reader, As you're reading this, I'll be on stage at the Econome conference in Cincinnati speaking about financial independence. (The talk will be on YouTube shortly.) Shout out to my friend, Diania Merriam I've also sent two issues of our new newsletter Future-Proof your Career with AI. I'll do wacky stuff like play with the $200 version of ChatGPT for a week to help you decide if it's worth it. Each week, this free newsletter will cover: Real-World AI Applications AI Tools & Tutorials...
Hey friend, I bit the bullet. Just for you guys. I wanted to see if ChatGPT pro is worth the absurd price I'd been stalling for a while. First, that's a lot of money. But more importantly, I didn't know how I could justify the ROI. What's included in Pro Some of the most vocal advocates of AI (like the economist Tyler Cowen) insist that if you're not using the latest models, you're missing out. The model I always hear people rave about is GPT o1 Pro - the most advanced reasoning model. So I...