I work with organizations on quantitative problems where standard data science isn't enough — forecasting in noisy, non-stationary environments; pattern detection in complex time-series; methodology development for

problems that don't fit off-the-shelf tools.

 

My background combines a PhD in computational sciences with industry experience in financial machine learning, predictive analytics for government regulators, and applied research at national institutes. I'm comfortable both with rigorous methodology and with shipping working systems.

​

How I can help

 

• Time-series forecasting and anomaly detection in scientific, financial, or operational data

• Predictive modeling and ML systems for regulatory or policy contexts

• Custom methodology development at the intersection of statistical physics, complex systems, and ML

• Applied research projects requiring rigorous quantitative methods

• Quantum computing applications, especially in optimization

• Data-driven analysis of complex networks and dynamical systems

 

Selected past collaborations

 

• UK Gambling Commission — predictive analytics for lottery-market regulatory analysis (via Pivigo S2DS, 2024)

• Turkish Airlines — ML-based oil-price forecasting (via TRK Technology)

• Ministry of Treasury and Finance, Turkey — macroeconomic forecasting models (via TRK Technology)

• TÜBİTAK National Metrology Institute — ML pipelines for the national early-earthquake-detection initiative (DETAM, ongoing)

 

How I work

 

I take on a small number of engagements per year, typically as:

 

• Short consulting engagements (a few weeks to a few months) for specific methodology questions or proof-of-concept work

• Research collaborations with academic, governmental, or industrial partners

• Advisory roles for projects where domain expertise in complex systems or data-driven science is needed

 

I'm based in Istanbul and work remotely or hybrid.