Co-Founder & CTO of I Know First Ltd. With over 35 years of research in AI and machine learning. Dr. Roitman earned a Ph.D  from the Weizmann Institute of Science

Highlights

• How AI helped Dr. Roitman overcome emotional bias and avoid the 2025 market crash
• A personal journey from chemistry to building a professional-grade predictive model for stock markets
• Why emotional discipline — not just data — is key in navigating turbulent financial markets

How I got saved from most of the market crash.

1. I’m not a rational trader, more of an emotional one. In my youth I started investing or you’d say trading. It was not very successful, to say the least. In the 1980’s I bought shares in a little known company Apple Computers. I sold them when I thought IBM was going to kill them.

2. In the early 1990s I bought my first real computer, a PC AT. When windows and MS office appeared I was among the first adopters, especially for excel. I was interested in investing, and hoped to develop an objective system to forecast stock prices. I started playing with the stock market data which became available for downloading by modem from special sites. That was before the internet.

3. I had some ideas how the markets interact, influencing one another like in a chaotic system. I built some models, which required optimizing coefficients using computer learning algorithms, but it was extremely slow on the machines available then. That was my hobby only, since I was working full time as a chemist. Some aspects of free radical polymer synthesis are very close conceptually to my understanding of how the stock market works. For my job I implemented several of my ideas for use in polymer chemistry. That was much easier because of the smaller number of variables to optimize.

4. At night I was still working on my models, getting more insights on how the market works. And my computer tools got faster every year. But it was only after I finally retired in 2007 I was able to fully commit to working on my prototype forecasting system. For that, one computer wasn’t enough, I had to develop a networking system of several computers. Gradually it was growing to what eventually became the I Know First forecasting system.

5. Now I’m still an emotional trader. But I have a tool at my disposal to counteract my emotions. Starting last December, I had conflicting emotions. I was reading the glowing forecasts in the press with stock prices rising, and watching the stark forecast tables getting red with downhill signals. I had terrible times deciding, until finally in February, I decided to pull the trigger. I sold most of my holdings in a few minutes before the major market plunge. Only after I did it did I finally feel relief.I’m good at many things, but not at trading, no matter how many tools I developed. Trading is too emotionally draining for me.

In stock trading, people often imagine success as a matter of intellect—cool-headed decisions, backed by charts, code, and models. But for many of us, especially those with tools and knowledge, trading isn’t just about logic. It’s about wrestling with ourselves.

It’s about emotion vs. reason.

When the Tools Say One Thing, But the World Says Another

I’ve built tools. Systems that scan, calculate, and forecast market behavior with impressive precision. One such tool had been warning me since December—flashing downhill signals in cold red cells while the headlines screamed euphoria.

Everywhere I looked, the press was glowing. Stocks were hitting new highs. Forecasts were optimistic, analysts smiling, and investors patting themselves on the back. It felt wrong to go against that wave.

But my tables—the ones I built—told another story. Quietly, methodically, they turned red. One by one.

The Emotional Battle Begins

That period was brutal. It wasn’t just indecision—it was emotional exhaustion. I knew what the data said. I could explain it in charts. I could model the probabilities. But acting on it? That’s where the storm was.

I hesitated. Days turned to weeks. Watching the market float higher while my confidence dropped lower. I began questioning myself. Was I seeing ghosts in the code? Was the market smarter than me again?

Then Came February

I couldn’t take it anymore. I finally pulled the trigger.

In a matter of minutes, I sold most of my holdings. It was fast, and it felt almost violent. Not because of the action itself—but because of the emotional toll leading up to it.

And then… relief.
Not euphoria. Not pride. Just a strange, deep calm. Like finally putting down a weight I’d been carrying for months.

Soon after, the market plunged—just like the signals warned.

I Know My Strengths, and I Know My Limits

I’m good at many things. I can analyze, build tools, and write logic that outperforms headlines. But I’ve learned something that’s hard to admit:
I’m not built for trading.

Not because I don’t understand it—but because I feel too much.
Because despite every signal and system, there’s still a voice in me that gets swept up in the crowd, in the press, in the fear of being wrong.

Trading Is a Test of Self

People say trading is a skill. That’s true. But it’s also a mirror. It reflects everything we are—our fears, biases, hopes, insecurities. The market doesn’t just test your strategy. It tests your emotional honesty.

If you don’t know who you are, the market will tell you.

For me, it told me this: I can build the compass, but I’m not the one to steer the ship through the storm.