Research

What we study

Yazan Alpha Technologies is a research company first. The product is a by-product of a standing scientific program: can an autonomous system discover, test, and govern its own market hypotheses with real rigor?

Themes

Active research themes

Autonomous strategy discovery

How an AI can generate candidate strategies from live market observation — and how to keep the generation process honest about what is signal and what is noise.

Evaluation methodology

Falsification-first testing: replay fidelity, out-of-sample discipline, shadow comparison of strategy versions, and the preservation of negative results.

Market microstructure

How markets behave at the event level — liquidity, timing, and data quality — and what that implies for the difference between a backtest and reality.

Governance of autonomous systems

Authority separation, evidence gates, and fail-closed design for AI systems that act in the world: making capability expansion a governed decision, never a side effect.

Founder research

The research background behind the company

Alpha Yazan AI's evaluation discipline comes directly from the founder's applied research practice. Before markets, that practice included healthcare AI — retrospective, de-identified clinical-data research on publicly available research datasets, focused on rigorous evaluation and leakage-aware methodology. That work is research background only: it is not a medical product and makes no clinical claims.

External · research.yazanalpha.com Ibrahiem Hamed — Research Portfolio Publications of record, project write-ups, methodology notes, and CV — on the founder's dedicated research subdomain.
Stance

How we report research

  • Claims are limited to what our evidence supports; status is published in plain terms.
  • Negative results are preserved and inform future work — they are not failures to hide.
  • Paper and historical results are never presented as profitability.