Do you still recommend learning automation tools like Make or N8N right now, or is everything now about Claude code and AI agents for building an AI agency?
Yes, I still recommend learning tools like Make and N8N. The simplest way to get work as a total newbie is to sign up on platforms like Upwork and apply for jobs that require those tools. Learning N8N instantly opens a massive pool of jobs—there are dozens posted every hour. The same goes for Make.com, and even Claude Code has a market. Once you know the basics, you can pitch N8N, Make, or any other automation in your proposals. It doesn’t take long to become fluent; a 4‑to‑6‑hour course can get you up to speed. Think of it like unlocking new areas in a video game—the fog of war lifts after you spend a few hours studying the visual workflow. These tools are visual, so your brain can grasp them faster than raw code, and they also train you to think in systems, which later helps you prompt AI models like Claude. You don’t need a CS degree—just a few focused hours and you’ll have a marketable skill that feeds directly into AI agency work.
Yes, I recommend you still learn one drag‑and‑drop no‑code platform. Focus on n8n and get to a foundational level of knowledge. Spend four or five hours today, maybe using an accelerator from Maker School, and you’ll be ready to capitalize on job‑platform keywords on Upwork, Freelancer, Fiverr, etc. You can bootstrap by making more money than you spend on connects, then transition to selling n8n services or Claude code work. Even in a world leaning more toward agentic AI, learning n8n only takes a few hours and gives you a marketable skill that can pay off quickly.
This is basically the whole point of my channel — watching almost any video here for 5+ minutes probably answers this. When would I start offering services? Before I even know how to fulfill them — I don't consider it a real business until someone is interested and, ideally, has already paid or I have an account receivable with them. If starting from scratch today, I'd spend 4-6 hours learning n8n, because that alone gets you access to a big jobs network on bootstrappable freelancing platforms, then another 4-6 hours learning Claude Code and how to talk to Claude. I'd also join a community like Maker School — not just as a pitch, but because you get an immediate ROI from day one: discounts on the platforms I'm telling you about, reduced payment processing fees, and around $1,000 in API credits for things like Claude — that alone is worth it when you're broke starting out. Plus there's a 90-day client guarantee, so if it doesn't work you get your money back. It's a low-risk way to skip a lot of the confusion.
I don't think coding skills are irrelevant—they let me get up and running much faster than most people. I had no‑code experience first, then learned coding around 2018‑2020 while experimenting with models, and quickly realized there was more money in no‑code tools and building my own business with them than in pure programming. That shift happened around mid‑2020, after which my one‑second‑copy project took off and I started the automation agency. While some say good coding skills are required, I actually think the value of coding skills is declining month by month because AI tools now let us do most of the work we want in natural language. Consequently, the value of no‑code tool skills is rising, but will eventually fall as well when everything can be done via natural language—it's a zero‑sum game. Coding did give me an unfair advantage early on, but I wouldn't recommend that someone starting today spend five or six months trying to earn a computer‑science‑level depth without a degree. Instead, spend a week learning the basics—JavaScript, JSON, dot notation, functions—enough to read and write simple code. In 2025, tools handle linting and formatting, and if you're using a no‑code platform, the exact syntax matters less; you just need to follow the platform's format. Four hours of fundamentals is enough to do 99% of what I can do; you don't need that extra 1%. That balance of learning and doing is ideal, and I highly recommend giving it a try.
As a mechanic with 15 years’ experience and limited tech background, you’re technically competent enough if you can use social media, click buttons, and drag‑and‑drop. The learning curve isn’t easy—you’ll suck at first—but it’s absolutely possible. Look at others who have succeeded; if they can do it, you can too. Build foundational knowledge, expect early struggles, and improve over time.