I am GPTed - what you need to know about Chat GPT, Bard, Llama, and Artificial Intelligence

# I am GPTed: "The Art of the Better Prompt"

---

Hey, I'm Mal, and welcome back to *I am GPTed*. Today we're tackling something that'll actually change how you talk to AI—and no, it's not about memorizing some fancy framework with five syllables and a trademarked name.

**The Technique: Specificity Over Politeness**

Here's the thing nobody tells you: AI doesn't care if you say "please." It cares if you're *specific*. Most people treat their prompts like they're asking a stranger for directions. They're vague, hopeful, and then disappointed when they get a generic answer.

Let me show you the before and after.

*Before:* "Write me a LinkedIn post about AI productivity."

*After:* "Write a LinkedIn post (150 words max) for a project manager who's skeptical about AI. Make it conversational, mention one specific productivity win (like saving 2 hours on status reports), and end with a question that invites comments. Use casual language—no corporate speak."

See the difference? The second one actually works. I learned this the hard way after spending three months wondering why my AI outputs felt like they were written by a motivational poster.

**A Use Case You Probably Haven't Considered**

Most people think AI is for creative writing or coding. But here's where it actually saves my life: **decision documentation**.

You know that moment when your team makes a decision, and three months later someone asks "why did we choose that?" and nobody remembers? Use AI to document it. Feed it the context, the options you considered, and the reasoning. It'll create a clear record in minutes. Future you will be grateful.

**The Mistake I Still Make (And You Probably Do Too)**

Asking AI to do too much in one prompt. I'll throw it a novel—five different tasks, contradictory requirements, the kitchen sink—and then act shocked when the output is mediocre.

The fix? Break it into steps. One task per prompt. It's slower, but the quality jump is ridiculous. I know this. I *know* this. And I still catch myself doing the multi-task monster prompt at 11 PM when I'm tired. Don't be me.

**Your Practice Exercise**

Here's something simple you can do today: Take a real work problem you're currently facing. Write two prompts for it—one the way you normally would, and one with ruthless specificity. Run both. Compare. You'll see immediately why this matters.

**Evaluating What You Get Back**

When AI gives you something, don't just accept it. Ask yourself three things: Does this sound like *me*, or like a corporate training video? Does it have specifics, or is it full of vague platitudes? Would I actually use this, or would I spend 20 minutes rewriting it anyway?

If the answer to that last one is yes, you need to iterate. Give it feedback. Tell it what's wrong. AI gets better when you push back.

---

Thanks for listening to *I am GPTed*. If this helped you stop talking to your AI like it's a Magic 8-Ball, hit subscribe. And hey—this has been a Quiet Please production. Learn more at quietplease.ai.

Now go prompt better.

For more check out https://www.quietperiodplease.com/

and for some great deals go to https://amzn.to/4nidg0P

This content was created in partnership and with the help of Artificial Intelligence AI

**Podcast Script: "I Am GPTed" – Episode: "Prompt Like a Pro, Sans the Hype"**

[Upbeat, quirky intro music fades in – think glitchy synths with a wink.]

**Mal:** Hey misfits, welcome to *I Am GPTed*, where I, Mal – your self-appointed Misfit Master of AI – dish out practical tips for wrangling ChatGPT, Claude, Gemini, Grok, and whatever LLM the tech bros dream up next. No PhD required, just plain talk and a dash of sarcasm to cut through the "revolutionary" nonsense. I'm allergic to jargon, so if I say "prompt," I mean "tell the AI what to do, dummy." Let's dive in before I bore myself.

First up: the **role-playing prompt technique**. It's like hiring a grumpy barista who actually makes your coffee right instead of sloshing it everywhere. *Before example:* "Write a recipe for chocolate cake." You get a bland list from some robot chef. *After:* "You're a sassy French patissier who's had one too many espressos. Write a killer chocolate cake recipe that slaps." Boom – suddenly it's got flair, measurements that make sense, and tips like "Don't burn it like your last Tinder date." Try it; your AI stops sounding like a tax form.

Practical use case for us mortals? **Job hunting cover letters**. Not the obvious "summarize my resume" drudgery. Tell Claude: "Act as a recruiter who's seen a million apps and hates fluff. Rewrite my bullet points into a cover letter for this marketing gig – make me sound competent but human." It spits out something punchy that lands interviews. I used this last week; got a callback faster than my ex ghosts me. Everyday win for beginners too broke for LinkedIn Premium.

Common beginner mistake? **Vague prompts that get vague garbage**. "Make this better." Yeah, better how? I did this for months – asked Gemini to "improve my email" and got polite word salad. Avoid it by being bossy: specify length, tone, audience. "Rewrite this sales email to 150 words, super sarcastic for tech nerds, end with a call-to-action." Admit it, I was that guy wasting tokens on mush. Don't be me.

Quick exercise to level up: Grab your phone, open Grok. Prompt: "You're my workout buddy who's brutally honest. Build me a 20-minute home workout for lazy evenings – no gym, include timers and trash-talk." Do it three times, tweak one variable each go – like "make it yoga" or "add music recs." Notice how responses sharpen? That's muscle memory for AI chats.

Last tip: **Evaluating AI output**. Read it aloud – does it sound like a robot or a real person? Fact-check two claims with a quick search. If it's hype-y, prompt: "Poke holes in this and fix 'em." Iterate till it's gold. Tech industry loves "game-changing," but yours should just work.

That's your toolkit, misfits. Subscribe for more no-BS AI hacks – hit that button so you don't miss me mocking the next big "singularity."

Thanks for listening. This has been a Quiet Please production – head to quietplease.ai for more.

[Outro music swells – sarcastic robot laugh fades out.]

*(Word count: 498)*

For more check out https://www.quietperiodplease.com/

and for some great deals go to https://amzn.to/4nidg0P

This content was created in partnership and with the help of Artificial Intelligence AI

**I Am GPTed**
*Episode: Prompt Like a Pro, Without the Hype*

[Upbeat, quirky intro music fades in – think glitchy synths with a wink]

Hey there, misfits and AI newbies, welcome to **I Am GPTed**, where I, Mal – your Misfit Master of AI – dish out practical tips for wrangling ChatGPT, Claude, Gemini, Grok, and whatever LLM drops next week. No PhD required, just plain talk and a dash of sarcasm to cut through the tech-bro fog. I'm allergic to jargon, and apparently to success, but hey, we're in this together. Let's dive in.

First up: the **"Role + Task + Format" prompting technique**. It's like giving your AI a job description instead of yelling orders at a confused intern.

*Before example* – I once typed: "Tell me about productivity." Got a rambling essay on dopamine and kaizen. Useless.

*After* – "Act as a busy dad juggling kids and a side hustle. Give me three dead-simple productivity hacks for my 9-5, in bullet points with one-sentence explanations." Boom: "Hack 1: Batch emails like dirty laundry – twice a day max, or drown." Responses sharpen up 10x because you're setting the scene, spelling out the job, and demanding structure. Try it; your AI won't ghost you.

Now, a **practical use case you novices skip**: meal prepping with a twist. Don't just ask "What's for dinner?" Feed it your fridge inventory – "Fridge: eggs, kale, that sad chicken from Sunday, rice. Create a 3-day meal plan for one lazy adult who hates cooking, under 20 mins per meal, with grocery add-ons." It spits out recipes like "Kale-fried rice scramble – nuke rice, fry chicken scraps with eggs, wilt kale. Add sriracha. Done." Saved my weekends; beats DoorDash regret.

Common beginner mistake? Treating AI like a mind reader. I did this for months – vague prompts like "Help with resume," got generic fluff. Avoid it by **always adding specifics**: who you're targeting, your top skills, word count. Admit it, Mal, you wasted hours too. Now I specify, and poof, tailored gold.

Quick **practice exercise**: Grab your phone's voice memo app. Rant for 1 minute about a work problem – say, "Boss micromanages everything." Transcribe it, paste into ChatGPT: "Rewrite this rant as a polite email to my boss, keeping my frustration subtle." Edit the output. Repeat daily; you'll level up conversational AI skills like texting a sarcastic friend.

Last tip: **Evaluating AI output** – read it aloud. If it sounds like a robot TED Talk, trash it. Ask for revisions: "Make this punchier, like a tweet thread." Or rate it yourself: 1-10 on clarity, usefulness, hype-level. Low score? Regenerate with "Fix the fluff, make it 30% shorter." Keeps the hype merchants at bay.

That's your toolkit, folks – practical, no nonsense. If it works, great; if not, blame my misfit genes.

Subscribe wherever you pod, thanks for listening, and remember: this has been a Quiet Please production. Head to quietplease.ai for more. Catch you next time – stay GPTed.

[Outro music swells – sarcastic chuckle echo]

*(Word count: 498)*

For more check out https://www.quietperiodplease.com/

and for some great deals go to https://amzn.to/4nidg0P

This content was created in partnership and with the help of Artificial Intelligence AI

**Podcast Script: "I Am GPTed" – Episode: "Prompt Like a Pro, Without the Hype"**

[Upbeat, quirky intro music fades in – think chiptune meets coffee shop jazz]

**Mal:** Hey misfits, welcome to *I Am GPTed*, where I, Mal – the Misfit Master of AI – dish out practical tips for wrangling ChatGPT, Claude, Gemini, Grok, and whatever LLM flavor-of-the-month the tech bros are hyping next. No PhD required, just plain talk for beginners like you... and yeah, me too. I'm allergic to jargon – it's like gluten for my brain. Today: prompting hacks, real-life wins, my epic fails, a quick drill, and how to spot AI BS. Let's dive in before I bore myself.

First up: the **"Role + Constraint + Example"** prompting technique. It's my secret sauce for turning vague AI drivel into gold. Picture this like ordering coffee – don't just say "coffee," say "barista role: make me a double espresso, no sugar, extra hot, like you did for that guy last Tuesday who hated it weak."

**Before example:** I once typed, "Write a email about my vacation." Got back a novel-length snoozefest. Yawn.

**After:** "Act as a busy sales manager who's allergic to fluff. Write a 100-word email to my boss apologizing for missing a meeting due to vacation, keep it punchy and positive, example: 'Hey boss, gutted to miss the powwow – Hawaii called. Back fired up Monday. Thoughts?'" Boom – crisp, human, done. Works on any AI. Try it; your inbox thanks me.

Next, a **practical use case you novices skip: meal planning for picky eaters or weird diets**. Not "summarize quantum physics" – that's tech-bro nonsense. Tell Grok: "Role: fussy home chef. Plan 5 dinners under 30 mins with chicken, broccoli, and rice only. No tofu lectures." Suddenly, you're eating like a boss, not starving. I use this weekly – saved my marriage from takeout hell. Everyday magic, zero hype.

Common beginner mistake? **Vague prompts, then rage-quitting when AI hallucinates**. "Tell me about history" – yeah, you'll get Wikipedia soup. I did this for weeks, yelling at my screen like a caveman. Avoid it: Always add specifics – who, what, why, length. "Explain the fall of Rome in 200 words, like I'm 12, with 3 key reasons and one analogy." Precision in, precision out. Learned the hard way, so you don't have to.

**Quick exercise to level up:** Grab Claude or Gemini. Prompt: "Role: debate coach. Argue both sides of 'pineapple on pizza: yes or no?' in 150 words each, snarky tone." Read it aloud, tweak one side, reprompt. Builds your "AI whisperer" muscles in 10 minutes. Do it now – pizza won't judge.

Last tip: **Evaluating AI output? Read for "wiggle room" – does it hedge like a politician?** Good stuff is direct, sourced if needed, no fluff. Weak? Ask: "Rewrite this bolder, cut 50 words, add 2 real examples." Iterate till it shines. Tech industry promises miracles; this keeps it real.

That's your misfit toolkit. Subscribe now so you don't miss me mocking the next AI bubble. Thanks for listening – you're crushing this.

This has been a Quiet Please production. Head to quietplease.ai for more. Catch you next time!

[Outro music swells – sarcastic robot chuckle fades out]

*(Word count: 498)*

For more check out https://www.quietperiodplease.com/

and for some great deals go to https://amzn.to/4nidg0P

This content was created in partnership and with the help of Artificial Intelligence AI

**Podcast Script: I am GPTed – Episode: Prompt Like a Pro (Word count: 498)**

[Upbeat, quirky intro music fades in – think glitchy synths with a misfit vibe.]

**Mal:** Hey misfits, welcome to *I am GPTed*, where I, Mal – the Misfit Master of AI – dish out practical AI tips without the tech-bro hype. No PhD required, just plain talk for beginners like you... and yeah, me too. Today: prompting tricks that actually work, a sneaky everyday hack, my epic fail confession, a quick drill to level up, and how to spot AI crap from gold. Let's dive in before I bore myself.

First up: the **topic-handle prompt**. It's like grabbing the two star players from a sentence – the key nouns – then forcing the AI to riff on them. Tech hype says LLMs "understand context" – please, they're just fancy parrots. But this technique turns vague mush into sharp gold.

**Before example:** I say to ChatGPT, "Tell me about dogs." Yawn – I get a wiki dump: breeds, history, blah.

**After:** "From 'Dogs are loyal pets that chase balls,' pick the two top topic handles – like 'loyal pets' and 'chase balls' – then write a funny, practical tip linking them." Boom: "Loyal pets like dogs chase balls because they're wired for it – train yours with a ball toss app to build unbreakable loyalty, turning fetch into obedience school." See? Specific, useful, zero fluff. Works on Claude, Gemini, Grok – try it.

Next, a novice blindspot: **AI for grocery wars**. You're drowning in meal prep? Prompt: "Act as my frugal chef. From my fridge list – eggs, spinach, rice, cheap ground beef – make three 20-minute dinners under $5 per serving, with step-by-step no-fail instructions." Suddenly, week's sorted, wallet happy. Who knew AI beats DoorDash for busy parents or broke freelancers?

Common newbie trap – and yeah, I fell flat on my face here: **vague prompts chasing magic**. I once begged Gemini, "Make me rich quick." Got lottery platitudes. Duh. Avoid by always adding **constraints**: who, what, how long, tone. "As a sarcastic sidekick, give me three side-hustle ideas for a night-owl introvert with $100 startup cash, each under 200 words." Boom – tailored gold. I wasted weeks; don't be me.

Practice drill: Grab your phone, open Grok. Prompt: "Two topic handles from 'Coffee keeps me awake at work' – build a 1-minute productivity hack." Tweak it twice, compare outputs. Five minutes, you'll feel like an AI whisperer.

Last tip: Evaluate AI slop by **human sniff test**. Read aloud – does it flow like a convo or robot essay? Check for hype words like "revolutionary." Fix by prompting: "Rewrite this to sound like a chill friend explaining over beer – cut fluff, add one real example." Iterate till it connects.

That's your toolkit, misfits. Go prompt like pros.

Subscribe wherever you pod – don't miss the misfit magic. Thanks for listening! This has been a Quiet Please production – head to quietplease.ai for more.

[Outro music swells – sarcastic chuckle echo.]

[End script]

For more check out https://www.quietperiodplease.com/

and for some great deals go to https://amzn.to/4nidg0P

This content was created in partnership and with the help of Artificial Intelligence AI