Subscribe to Hard Mode First
>600 subscribers
Hard Mode FirstChatGPT Saved My Life (No, Seriously, I’m Writing this from the ER)
How using AI as a bridge when doctors aren't available can improve patient-to-doctor communications in real time emergencies
Hard Mode FirstHow to Plan an Annual Family Summit
Simple strategies for setting goals and Priorities with Your Partner for the year ahead
Hard Mode FirstHow I Used AI to Save My Life in 77 Prompts: A Debrief
Reflecting on best practices, lessons learned, and opportunities to improve AI-assisted medical triage
Subscribe to Hard Mode First
>600 subscribers
Hard Mode FirstChatGPT Saved My Life (No, Seriously, I’m Writing this from the ER)
How using AI as a bridge when doctors aren't available can improve patient-to-doctor communications in real time emergencies
Hard Mode FirstHow to Plan an Annual Family Summit
Simple strategies for setting goals and Priorities with Your Partner for the year ahead
Hard Mode FirstHow I Used AI to Save My Life in 77 Prompts: A Debrief
Reflecting on best practices, lessons learned, and opportunities to improve AI-assisted medical triage
Share Dialog
I’ve started getting a very specific kind of request when companies ask for AI 101 trainings.
The questions usually sound like this:
“Which LLM (ChatGPT, Claude, Gemini) and models should we use for each task?”
“What’s the best tool for image generation? Video creation? Slide deck creation?”
“What are all of the ways you can use this particular tool?”
While these are practical questions, leading with a tools-first mindset is like teaching someone how to make a quiche by giving them a tour of your industrial-grade kitchen.
At a certain point, the rundown of every utensil and appliance becomes noise.
Let’s face it: After a single kitchen tour, you’re never going to remember which knife does what. And honestly? That’s fine. You don’t need to know what every knife does. At least, not yet.
You just need to know how to dice the onion you need for your quiche.
The tool matters less than the problem you’re trying to solve.
Share Dialog
I’ve started getting a very specific kind of request when companies ask for AI 101 trainings.
The questions usually sound like this:
“Which LLM (ChatGPT, Claude, Gemini) and models should we use for each task?”
“What’s the best tool for image generation? Video creation? Slide deck creation?”
“What are all of the ways you can use this particular tool?”
While these are practical questions, leading with a tools-first mindset is like teaching someone how to make a quiche by giving them a tour of your industrial-grade kitchen.
At a certain point, the rundown of every utensil and appliance becomes noise.
Let’s face it: After a single kitchen tour, you’re never going to remember which knife does what. And honestly? That’s fine. You don’t need to know what every knife does. At least, not yet.
You just need to know how to dice the onion you need for your quiche.
The tool matters less than the problem you’re trying to solve.
Maybe you’re an entrepreneur trying to diversify your customer base into a more vertical offering. Maybe you’re a sales leader looking to streamline how you gather context on each account. Maybe you’re a mom who just wants to motivate your kid to practice the piano.
Pick a problem. Get clear on what success would feel like on the other side. Then decide which AI tools can help.
I know this seems like a counter-intuitive approach to learning, particularly for those of us who grew up in a textbook generation of learning a new concept from start-to-finish, in order.
But learning AI is a lot less like learning how to learn a tool and a lot more like learning how to learn a language.
Learning a language requires three things:
Basic vocabulary and grammar (the conventions)
Practice on your own (reading, writing, speaking)
Practice in a group (immersion)
It’s nearly impossible to learn an entire language with simple vocab-memorization. Those among us who pick up languages the best are people who are lucky enough to embed themselves in those countries or setting. You build fluency through immersion.
Learning AI is the same way.
Sure, it’s helpful to have a few guardrails and signposts, particularly among things like how to keep your data secure, and where AI might not be the best tool for the task. But aside from that, you just need to get in the weeds and play.
There’s a reason most of us learn how to cook by following a recipe, either from a cookbook, a video, or someone standing next to us in the kitchen.
Making a dish teaches you the end-to-end process of what’s required to cook something new. You have a clear beginning, middle, and end. Each recipe requires that you use some utensils and ingredients along the way. And after a few rounds, you start to recognize the pattern. You increase the complexity. You improvise.
Over time, you develop your own style. Maybe you discover an affinity for baking cakes. So you start to specialize. You invest in a KitchenAid stand mixer, you watch a few YouTube tutorials on frosting best practices. Soon enough, you’re the one your neighbors are asking to bake cakes for all of their kids’ birthday parties.
Learning AI works the same way.
You don’t start by mastering every model, framework, or tool. You start by building one useful thing. Maybe it’s a workflow, a prompt, or a small automation that solves a real problem you actually have. Through that process, you learn which tools matter, when they help, and when they don’t.
Over time, you develop a style. Maybe you discover you’re great at building internal tools. Or automating research. Or turning messy inputs into clear outputs. So you go deeper. You invest in the tools that support your work. Others start coming to you for help.
That’s what fluency looks like.
You don’t get it from a single workshop, or a tour of the kitchen. You get it by giving people permission to build.
Get your laptops out. It’s time to cook.
Maybe you’re an entrepreneur trying to diversify your customer base into a more vertical offering. Maybe you’re a sales leader looking to streamline how you gather context on each account. Maybe you’re a mom who just wants to motivate your kid to practice the piano.
Pick a problem. Get clear on what success would feel like on the other side. Then decide which AI tools can help.
I know this seems like a counter-intuitive approach to learning, particularly for those of us who grew up in a textbook generation of learning a new concept from start-to-finish, in order.
But learning AI is a lot less like learning how to learn a tool and a lot more like learning how to learn a language.
Learning a language requires three things:
Basic vocabulary and grammar (the conventions)
Practice on your own (reading, writing, speaking)
Practice in a group (immersion)
It’s nearly impossible to learn an entire language with simple vocab-memorization. Those among us who pick up languages the best are people who are lucky enough to embed themselves in those countries or setting. You build fluency through immersion.
Learning AI is the same way.
Sure, it’s helpful to have a few guardrails and signposts, particularly among things like how to keep your data secure, and where AI might not be the best tool for the task. But aside from that, you just need to get in the weeds and play.
There’s a reason most of us learn how to cook by following a recipe, either from a cookbook, a video, or someone standing next to us in the kitchen.
Making a dish teaches you the end-to-end process of what’s required to cook something new. You have a clear beginning, middle, and end. Each recipe requires that you use some utensils and ingredients along the way. And after a few rounds, you start to recognize the pattern. You increase the complexity. You improvise.
Over time, you develop your own style. Maybe you discover an affinity for baking cakes. So you start to specialize. You invest in a KitchenAid stand mixer, you watch a few YouTube tutorials on frosting best practices. Soon enough, you’re the one your neighbors are asking to bake cakes for all of their kids’ birthday parties.
Learning AI works the same way.
You don’t start by mastering every model, framework, or tool. You start by building one useful thing. Maybe it’s a workflow, a prompt, or a small automation that solves a real problem you actually have. Through that process, you learn which tools matter, when they help, and when they don’t.
Over time, you develop a style. Maybe you discover you’re great at building internal tools. Or automating research. Or turning messy inputs into clear outputs. So you go deeper. You invest in the tools that support your work. Others start coming to you for help.
That’s what fluency looks like.
You don’t get it from a single workshop, or a tour of the kitchen. You get it by giving people permission to build.
Get your laptops out. It’s time to cook.
1 comment
I’ve started getting a very specific kind of request when companies ask for AI 101 trainings. Questions like: - “Which LLM (ChatGPT, Claude, Gemini) and models should we use for each task?” - “What’s the best tool for image generation? Video creation? Slide deck creation?” - “What are all of the ways you can use this particular tool?” While these are practical, leading with a tools-first mindset is like teaching someone how to make a quiche by giving them a tour of your industrial-grade kitchen. At some point, it doesn’t really matter whether you know what each knife does. You just need to know how to dice an onion. Learning AI works the same way. You don’t start by mastering every model, framework, or tool. You start by building one useful thing. Maybe it’s a workflow, a prompt, or a small automation that solves a real problem you actually have. More on this in today's post... https://hardmodefirst.xyz/to-learn-ai-skip-the-tour-of-the-kitchen