Yesterday I wrote about going from lost to building. Zero code, a budgeting app for my wife, a company university, a deal hub. A few of you replied with some version of the same question. That is great, but where do I actually start.

So here is the how. Not theory. The exact ladder I climbed, and the one thing I would tell you to do this weekend.

First, three things that were holding you back

Before any tool, the mindset. These are the ones that unstuck me.

You are not behind. AI rewards domain expertise more than time spent with the tools. Your years in GTM are the edge here, not the handicap. The 22-year-old who knows the tools but has never carried a number is not ahead of you. You just have to point what you know at the tool.

It is a coworker, not a search box. A search engine is something you query and close. A coworker is someone you onboard once, and it compounds every day after. Most people use the most powerful coworker they will ever have like a slightly better Google. Stop.

Fifteen minutes a day on real work beats a weekend course. Not textbook examples. Your actual pipeline export, your actual account, your actual messy problem. Reps on the real thing is the only thing that moved me.

The ladder

You do not need all of this. You need the next rung.

1. Set it up as a coworker. Give your AI standing instructions once, so you are not re-explaining yourself every chat. Tell it who you sell to, your positioning, your role, and how you want it to talk to you. Direct, no flattery, ask before assuming.

2. Learn the one prompt shape. Persona plus task plus context plus format. "You are a senior RevOps analyst. Turn this pipeline export into a five-bullet risk summary for my VP. Data: [paste]. Lead with the biggest risk." That covers most of what you will ever need. And kill the yes-man. Tell it to challenge you, or it will just agree with everything you say.

3. Connect your tools. This is the unlock most people never reach. Through something called MCP, you can plug your real tools into your AI so it reads and acts on your actual data instead of guessing. You connect it, you do not build it. Think USB-C for AI.

4. Build one small thing. This is where it got fun for me. Budget B started because my wife hates how every budgeting tool overcomplicates something simple. I described what she wanted in plain English and the tool wrote the code. That one small win cracked everything open. Pick something equally small and equally annoying.

5. Make it repeat. Once a chat does something useful twice, turn it into a reusable skill or a scheduled automation. That is the difference between a neat trick and a system that works while you sleep.

Where to go next

I put the full version of this, with the exact starter prompts and a short exercise for each rung, in one place.

The field guide. Every step above, expanded, with prompts you can paste and a real exercise for each. Start here: builtgtm.ai/how-to-ai

Which of your tools connect, and how. A plain-English directory of what plugs into your AI and the exact steps for each: builtgtm.ai/stack

Something to copy. I built a job-search engine that scores roles against what I actually want, so I stop drowning in noise. The whole thing is written up, free: builtgtm.ai/playbooks/job-engine

Ready-made systems. The workflows I run, packaged so you can install and run them yourself: builtgtm.ai/tools

Your one assignment this weekend

Pick one task you dread. The report you put off, the data you never look at, the message you rewrite five times.

Give it to your AI. Spend fifteen minutes. Do not try to automate your whole job. Just do that one thing, once, with a coworker instead of alone.

It is amazing what your own data will tell you once you get a little curious and know how to ask. That curiosity is the only prerequisite. Everything else is on the other side of the door.

I know zero code. If I can build from here, so can you.