This told essay is based on a conversation with Yesim Saidan, a branding and communications expert in his early 50s based in the Netherlands. The following has been edited for length and clarity.
When you’re stuck on a business decision or need to come up with a creative idea or strategy, brainstorming usually starts with Steve Jobs’ Custom GPT.
My private consulting business helps senior executives and high-profile entrepreneurs increase their authority and influence through social media and brand strategy. But scaling that work on my own was difficult.
Before AI, if you wanted to increase the number of clients you could take on, your main option was to hire freelancers for special projects and tasks. I spent a lot of unnecessary time training freelancers on my particular framework, and often felt like they didn’t care as much as I did.
Everything changed when OpenAI launched a custom GPT. We’ve created over 17 custom GPTs and built our team using this feature. Next, I thought of my ideal mentor and created a custom GPT for that mentor.
I had to create 4 or more custom GPTs to get good results
One of my first jobs was working as a Wall Street project manager at Citibank. After immigrating to the United States from Turkey for an MBA. This was the beginning of my 14-year corporate career, during which I worked in New York, Paris, and the Netherlands.
I started my business about 10 years ago because I wanted more flexibility in my work schedule. At the time, social media was just starting to take off, and I saw a clear opportunity.
We had used AI tools before, but OpenAI’s custom GPT was a game-changer. Initially, I envisioned my ideal team of four agents. We quickly realized that if the AI was overloaded with too many tasks, it would produce substandard results.
Instead, we created custom GPTs for each important task we wanted the AI to perform. So now we have over 17 custom GPTs that feel like the perfect team.
We trained our AI team to focus on the bigger strategy
You can create a custom GPT in 5 to 10 minutes, but what really makes it powerful is the training process. I create standard operating procedures for each task and client and use them as training materials for agents to outline my methodology and framework.
I have client-specific AI agents trained based on each major client’s tone, goals, and past conversations, so I never start a task from scratch. Training is ongoing. Every time you create a query or upload a document, your agents improve just like real employees.
When you need to communicate or create content in the client’s tone, the final draft is so customized that it feels like you spent hours perfecting it, when in reality it was handled by an AI team.
I’ve trained market researchers, cold call analysts, proposal writers, video scriptwriters, and even custom GPTs to evaluate LinkedIn profiles using six pillars to determine whether your current LinkedIn presence is building authority, attracting your ideal customers, and establishing authenticity, clarity, and uniqueness. These have allowed me to focus on the big picture strategy.
We taught custom GPT to think and lead like Steve Jobs
After creating my ideal employee, I asked myself which mentor I would want living or dead. Steve Jobs is known for his creativity and innovation, and numerous videos about him have already been published online. He is the perfect mentor for creating custom GPTs.
The instructions started by saying something like, “You are Steve Jobs. You have decades of experience in X, Y, and Z. Your most important skills are creativity, innovation, and thinking outside the box.”
There are two types of video transcripts that I trained on. I’ve uploaded a video transcript where he explains his strategy and what he looks for in a product. The second approach is training with examples. We found videos showing how he launched products like the iPhone and iPad, so the AI learns from both his thought process and the execution of those launches.
To reach our current level, we spent approximately 40 hours researching and building training assets, including PDFs and other materials that GPT can use as reference. I continue to add relevant material as I find it, and I’ve now even created custom GPTs for Dan Kennedy and Elon Musk.
Need to avoid certain questions to get the most appropriate answer
The frustrating thing about training an AI model is that you can give it a lot of the information it needs to perform Steve Jobs’ superpowers, but the AI can take it and produce different things.
When encouraging questions, avoid questions like, “What do you think about this idea?” Because AI usually agrees with me and wants to please me. Instead, ask, “On a scale of 1 to 10, how good is this idea?”
I’m not saying this idea is bad, but you might say it’s a 5 now. Then ask, “Okay, what would make a 10?”
Usually that’s when I start leveraging the Steve Jobs experience I’ve been training with. Iterate over and over until you get the most helpful and honest feedback possible.
It depends on the task. It typically takes 3-5 adjustments to create a more strategic deliverable.
Continue reading the Tiny Teams series
AI is scary, but there’s no going back
When products like NotebookLM came out, I started thinking, “Oh my god, this is going to make the entire human race obsolete.” AI products can be fascinating at first, but they can also be really scary.
I really think that we don’t know what the world will be like even a year from now. Sometimes I literally freeze thinking about the effects and how everyone will end up homeless, but usually I try to remind myself that I’m not God or a higher power and I don’t know what’s going to happen. This will calm you down.
We also realized that while AI is powerful on its own, it’s when we combine our expertise and skills with it that it becomes truly magical. When you use custom agents as an extension of your brain, rather than a replacement for it, you can achieve really great results.
There is no turning back from there.
