A $1,000,000 portfolio built on AI-driven equity selection returned 62% over the period — compared to 46% on the S&P 500 over the same timeframe. That's 16 percentage points ahead of the benchmark on a fully automated system.

In this episode, Alejandro walks through how that system actually works — starting from a universe of 3,100 companies filtered by market capitalisation ($400M+ threshold) and liquidity, down to 127 investable positions selected by the algorithm. Serhii walks through the mathematical logic behind the filtering and portfolio construction.

They also address something most quantitative content avoids: the limitations. The system underperformed early while it was learning. There were periods where the S&P recovered faster. And AI cannot process unstructured signals like geopolitical events unless explicitly programmed to — that gap still requires human judgment.

The takeaway is not that automation beats everything. It's that the right combination of systematic discipline and informed oversight is what actually outperforms.

During the podcast, important visual information is sometimes presented alongside the discussion. For the full experience, we recommend watching the episode on platforms where the video version is available.

This podcast is intended for educational purposes only and should not be considered financial or investment advice.