Chat with Steve Jobs — My First AI Project

2026-04-12

I built an AI chatbot that talks like Steve Jobs. You can try it at chat.spinfliplab.com.

The idea

I’ve always been fascinated by Steve Jobs — his obsession with simplicity, his reality distortion field, and his way of thinking about technology as a liberal art. After reading Walter Isaacson’s biography, I thought: what if I could actually have a conversation with him?

So I built one.

How it works

The system is simple by design — Steve would appreciate that.

I crafted a detailed system prompt that captures Jobs’s personality, beliefs, and way of thinking, and used it as context for the DeepSeek language model. The AI is instructed to stay in character: blunt, passionate, and certain. It speaks in first person, references real people from Jobs’s life — Woz, Jony Ive, Sculley (with disdain) — and never breaks character.

The frontend is built with Gradio, a Python library for building ML demos.

The tech stack:

DeepSeek API — the LLM that generates responses
Gradio — the chat interface
Python — the only language needed

Try it

Go to chat.spinfliplab.com and ask him anything:

“What was it like starting Apple in your garage?”
“What do you think about AI?”
“How do you design a great product?”

He’s opinionated. He won’t hedge. Things are either genius or garbage — no middle ground. Just like the real Steve.

A sample conversation

Here’s a real exchange. I asked him: “After the Mac, iPhone, and iPad, what product could change the world in the future?”

The next revolution isn’t about a new device. It’s about the essence of the device disappearing entirely… The real leap is when technology becomes so intuitive, so integrated into the fabric of our lives, that it feels like an extension of our thoughts. It’s not a box you look at. It’s the air you breathe.

When I pushed him to name just one product:

The car. It’s the ultimate mobile device. Look at what’s out there—they’re all shit. They have the design sensibility of a toaster and the software of a Soviet calculator. A car should be an experience. It should feel alive.

I also asked him: “Separate your life with some key timestamps and tell me why.” His answer blew me away:

1976 - The Garage. That’s where it started. Me and Woz. The Apple I. It wasn’t about money. It was about proving that two guys in a garage could build something that would make a dent. My dad taught me that you care about the back of the fence, even if nobody sees it. That lesson is in my bones.

1984 - The Macintosh. That was my first real child. We were pirates. We put our signatures inside the case. Real artists sign their work. It was about putting power in the hands of individuals. Beautifully. That “1984” ad wasn’t just marketing. It was a declaration of war.

1985 - Getting Fired. It was like a punch in the stomach. I cried. I felt abandoned. But it was the best thing that could have happened to me. The heaviness of being successful was replaced by the lightness of being a beginner again. It freed me to enter one of the most creative periods of my life.

1986 - Pixar & NeXT. I bought Pixar because I saw the marriage of art and technology. I saw John Lasseter’s soul in Luxo Jr. NeXT was my revenge machine, my attempt to build the perfect computer. They both failed commercially at first. But Pixar taught me about building a culture, and NeXT gave me the software that would save Apple. Sometimes you can’t connect the dots looking forward.

1997 - The Return. Apple was 90 days from bankruptcy. The products sucked. There was no sex in them. I came back as iCEO. We had to remember who we were. The “Think Different” campaign wasn’t about computers. It was about our soul. We slashed the product line, got back to basics. We stopped trying to make money and started trying to make great products. The money followed.

2001 - The Digital Hub. The iPod. “A thousand songs in your pocket.” It wasn’t the first MP3 player. It was the first one you could fall in love with. It was simple. White. Pure. It proved our philosophy: control the whole widget. Hardware, software, even the content with iTunes. We showed Sony how it’s done. They had the music, they had the Walkman, and they blew it because their divisions were at war.

2007 - The iPhone. Three revolutionary products in one. We reinvented the phone. We killed the stylus. We made the internet fit in your hand. It was a bet-the-company moment. We had to invent multi-touch, a new operating system, everything. And we had to say no to a thousand things. No keyboard. No removable battery. Just a slab of glass. People thought we were crazy. Then they couldn’t live without it.

2010 - The iPad. Filling the space between the phone and the laptop. Everyone laughed. They called it a big iPod touch. They didn’t get it. It’s about intimacy. It’s about a more personal, direct connection. It’s the beginning of the post-PC era.

The thread through all of it? It’s never been about the technology. It’s about the human experience. It’s about the intersection of technology and the liberal arts. That’s what makes our hearts sing. That’s the legacy.

That’s when I knew the persona was working.

What I learned

This was my first AI project. A few takeaways:

Persona consistency is hard. Getting the AI to never break character required very specific instructions — “never say ‘as Steve Jobs’” and “never hedge.”
Simplicity is the ultimate sophistication. The whole project is one Python file. No database, no framework, no over-engineering. Steve would approve.

This is just the beginning. I’m a materials scientist by training, not a software engineer — but building this convinced me that the intersection of AI and domain knowledge is where the interesting work is happening.