Knowing how to use AI is only half the battle. A tool with nothing to work on just sits there. This part ties the series together: how to find something worth pointing AI at, something you care about, know well, and that directly benefits you.
Half the battle
Many people are now very good with AI. They write sharp prompts, build custom assistants, and set up agents that run on their own. Yet a lot of that skill goes unused, because they have nothing specific to point it at.
Think of a fully equipped workshop. The best saws and drills in the world don't build anything without a project. AI is the same. Skill with the tool is half the battle. The other half is the target: a goal, a problem, a product, or a business that the AI can work on every day.
The best targets share one trait: they come from you. That fits the idea from Part 0. AI can execute, but the new idea, the problem you see, and the reason it matters have to come from a person.
Finding your target
A good target answers "yes" to four questions. The more yeses, the longer you'll stick with it, and the better the AI's work will be, because you can judge it.
| Question | Why it matters | Ask yourself |
|---|---|---|
| Do I care about it? | Passion keeps you going when progress is slow | What do I read about or talk about for fun? |
| Do I know it well? | You can catch AI's mistakes and add what it doesn't know (Parts 1 and 3) | Where have I spent years of work, practice, or experience? |
| Does it benefit me directly? | Time saved, money earned, or a problem in my own life solved | What eats my time, or what would I pay to fix? |
| Is there a real problem? | Someone, maybe you, has a pain that existing answers don't fix | What do people around me keep complaining about? |
Prompts to help you find it:
- "Here's my background: [work, hobbies, skills]. List 10 problems people in these areas complain about that don't have a good solution yet."
- "I spend most of my week on [tasks]. Which of these could AI do for me, and what would I do with the time I get back?"
- "Here are three ideas I'm considering: [ideas]. For each, tell me who already solves it, and who would pay for a better answer."
Then use the two-sides prompt from Part 1 on your favorite idea before you commit to it.
Working your target, step by step
Once you have a target, work it with AI like a partner, not a search box. This is a simplified version of the process I use.
- State the goal in one sentence. "Earn extra income from my baking without working more hours."
- Try the obvious answer first. Ask AI for the standard solution and see why it falls short. ("Sell at the farmers market": that takes every Saturday, so it breaks the goal.)
- Turn the failure into rules. "No weekend shifts. No large upfront cost."
- Pick the real problem. Use your own experience. "Local cafes complain they can't get fresh gluten-free pastries."
- Check who already solves it and who pays. Ask AI to search for existing products and businesses, and for buyers already spending money on this pain.
- Break it down and improve it, round after round. Ask: "What's the hardest part of this? How could we make it simpler?" Repeat until the idea stops changing. Expect about ten rounds.
- Ask other AIs to attack it. Paste the idea into a fresh chat or a different tool: "Find every weakness in this plan." Keep what survives.
- Run the smallest real test. One cafe, one week, one batch. A real result beats any amount of planning.
The AI does the research, drafting, and number crunching at every step. You supply the goal, the problem, and the judgment of "did this get better?" That's AI executing and you creating.
Why I built the Forge
This is the gap I built Raya.ai to close. The process above came out of a long back-and-forth with AI that led me to Project Crystalline, a software idea for easing strain on the power grid. Afterward I realized the process mattered more than the single idea. If it worked for me, it could work for anyone with something they know and care about.
The Forge in Raya.ai is designed to turn that process into steps anyone can follow. You bring the goal, the problem you see, and what you know. The Forge helps you shape it into a real product or service, something a specific buyer would pay for. Your Raya agents then automate most of the research, outreach and follow-ups. You decide what goes out. Nothing is sent, signed or paid without you.
That's the difference between an agent that sits idle and one that works for you.
You can do everything in this series on your own with any AI tool. If you want something that does it all for you, personalized to you, and built around the one thing AI can't reproduce or recombine on its own, which is you, that's why Raya.ai exists.
Follow the journey. I'm proving this thesis in public by building and selling Crystalline, my first product to come out of the Forge. Follow along on LinkedIn (opens in a new tab), YouTube (opens in a new tab), X (opens in a new tab), and Reddit (opens in a new tab), or join our Discord (opens in a new tab).
Found this helpful? Raya.ai is built on the ideas in this series. Learn more at raya.ai (opens in a new tab), or download the app on iOS (opens in a new tab) or Android (opens in a new tab).