That Book Thick Enough to Hammer Tent Stakes
In middle school, I wanted to learn to code.
Went to the bookstore and bought a Visual Studio book. Brought it home, opened chapter one — and that was the end of that.
That book was thick enough to use as a hammer for tent stakes.
From then on, I never touched code again. I went down a completely different path — financial advising. Helping people with financial planning, risk analysis, communicating needs.
Over a decade later, I single-handedly built with AI: a brand website, a Threads automation system, an AI security scanner, and an Agent Fleet automation fleet.
I still can't code.
What Does "Can't Code" Actually Mean?
Let me be clear: I'm not saying code isn't important. Code is the skeleton of a product. Without code, nothing runs.
What I'm saying is: You don't need to write it yourself.
In 2026, AI's coding ability has reached a point where — you tell it what you want, it writes it. You check if it works, ship it if it does, tell it what broke if it doesn't.
How much code do I actually understand now?
Honestly, I only understand "it works" and "it doesn't work."
My debugging process looks like this:
Press F12 → Look for red errors → Copy the error message → Feed it to AI
That's it. Nothing more.
And that's enough.
Three Things More Important Than Coding
If coding isn't important, what is?
1. Communication Skills
This is the most important skill I learned from financial advising.
What's a financial advisor's daily work? A client says: "I want to save money." Then you need to figure out what they really mean — save for retirement? Buy a house? Or just feeling anxious?
You take a vague need and break it down into a concrete plan.
Communicating with AI is exactly the same.
"Make me a website" is a vague need. You need to turn it into: "Make me a single-page brand website, dark background, with service listings and a contact form, built with React and deployed to Vercel."
That's not a programming skill — that's a communication skill.
2. Understanding Purpose
All my products grew from my own needs.
MindThread (Threads automation system) — I was already posting on Threads, but the official scheduler sucked. So I built my own.
UltraAdvisor (advisor brand page) — I needed something to show clients, solving the "sales reps don't know what content to post" and "financial visualization" problems.
Every product's starting point wasn't "I want to learn a technology" — it was "I have a problem to solve."
Technology is the means. Purpose is the direction.
You don't need to know how React's Virtual DOM works. You need to know: "What problem am I solving? What should the end result look like?"
Think that through, and AI handles the rest.
3. Enjoying the Exploration Process
Building products isn't linear. You don't follow a textbook from chapter one to the last chapter and — done, product complete.
The real process looks like this:
Have an idea
→ Chat with AI to see if it's feasible
→ Try building the simplest version
→ It breaks
→ Fix it
→ Discover you can add a feature
→ Add it, breaks again
→ Fix it
→ Ship it
→ Users say this part sucks
→ Fix it
→ Breaks again
→ ...
If you can't handle this process, no coding course will save you.
But if you enjoy this "explore → try → fix" cycle, you don't need courses at all. AI is your best teacher and partner.
"Should I Take a Coding Course?"
If someone asks me this, I'd first ask back:
"What do you want to create?"
If your answer is "I want to learn Python" — you probably don't need a course. What you need is a problem to solve. Languages are tools, not goals.
If your answer is "I want to build a tool that auto-schedules Threads posts" — you can start right now. Open AI, tell it what you want.
Coding courses teach syntax. But in the AI era, syntax is the least valuable thing.
What's valuable:
- Can you articulate your requirements clearly
- Can you judge if the result is correct
- Can you find the problem when things break
These three skills can't be learned in a coding course.
As Your Portfolio Grows, You'll Naturally Understand
An interesting phenomenon: I've never formally "learned" any programming language, but as I build more things, I've started understanding some code.
Not because I took a course. Because every day I look at code AI writes — how it solves problems, how it organizes architecture. Look at enough, and you develop instinct.
It's like you don't need music theory to tell if a song sounds good. Listen to enough music, and you develop judgment.
You don't need to write code. You need to read code.
And the standard for "reading" isn't understanding every line of syntax. It's being able to judge after looking: Is this thing doing what I wanted?
Build a few products, and you'll have this ability.
The Unexpected Advantage of a Finance Background
Many people think a "non-technical background" is a disadvantage. But my finance background actually gave me several advantages that technical people might not have:
Cost awareness — For every technical decision, my first thought is "How much does this cost?" Can the free tier handle it? When is it worth paying? This mindset helped me build complete products on a $0 budget.
Risk awareness — API Key leak? Pushing to production without testing? In a finance person's eyes, these are all risk management problems. Think worst case first, then decide whether to proceed.
Client thinking — Products aren't built for self-satisfaction. Who will use it? What problem does it solve? Will they pay? Financial advisors ask these questions every day.
Your background isn't your limitation. It's your differentiation advantage.
The Times Have Changed
Over a decade ago, that Visual Studio book represented an era: Want to build tech products? First spend three years learning syntax.
In 2026, the rules changed.
You don't need three years before starting. You can start today.
You don't need to understand underlying architecture. You need to understand the problem you're solving.
You don't need to know how to code. You need to know how to communicate with AI.
That book thick enough to hammer tent stakes — I'm glad I gave up on it. Because if I'd forced myself through it back then, I might have become an ordinary engineer.
Instead of a financial advisor who built five products with AI.
For People "Wanting to Learn to Code"
If you're currently debating whether to learn coding, ask yourself three questions:
1. What do I want to create?
→ If you can't answer, find a problem first, not a course
2. Why do I want to learn coding?
→ If the answer is "feels like I should know it," you don't need to
→ If the answer is "I want to build XX but don't know how," just open AI and build it
3. Do I have a problem I want to solve?
→ Yes → Open AI, start building
→ No → Go find one in your daily life first
Coding is the means. Creating is the purpose.
Don't mistake the means for the purpose.
This is part four of the "Getting Started" series. Previous: AI Development Pitfall Diary: Mistakes I Made So You Don't Have To
Want more free resources? Join the Solo Lab Discord.
Discord: https://discord.gg/ewS4rWXvWk
Originally published on Ultra Lab — we build AI products that run autonomously.
Try UltraProbe free — our AI security scanner checks your website for vulnerabilities in 30 seconds: ultralab.tw/probe
