Alea
Nondum



Alea Nondum is a personal applied laboratory for turning forecasting ideas into a running system: ingesting market and weather data, producing forecasts, storing results cleanly, and presenting them in a simple public interface.

The project is deliberately end-to-end. It is not only about the model, but also about the engineering patterns needed to operate a forecasting workflow continuously and make its output inspectable.

It is also an experiment in AI-assisted engineering, but not a vibe-coding project. Part of the motivation is to explore the practical trade-off between delegating work to AI tools and remaining in the driving seat: building and deploying faster without giving up architectural ownership, code understanding, or the learning experience.

Dr. Raffaele G. Sgarlato

Energy Software & Machine Learning Engineer

Forecasting · Optimization

Berlin, Germany

Demo

This forecast explores a data-parsimonious approach to German day-ahead electricity price modeling. Instead of depending on external load or renewable generation forecasts, the model works directly with recent prices, commodity context, calendar features, and raw ECMWF AIFS numerical weather forecasts.

Attention Mechanism

Methodologically, the interesting part is temporal alignment. Price observations, market fundamentals, and weather forecasts live on different grids, with different horizons, publication rhythms, and information sets. The model uses learned time encodings and attention mechanisms to relate discontinuous price history, forecast weather fields, and the target delivery window without forcing every signal into a single hand-engineered interpolation scheme.

Limitations

At this stage, the live forecast is a point-forecast baseline: deliberately lean, inspectable, and useful for exercising the full data-to-forecast pipeline. One area to refine is the averaging and smoothing tendency of the attention mechanism, to capture sharper price movements and leverage autoregressive behaviour more effectively. Another direction I am considering is probabilistic forecasting, for example productionizing an ensemble prototype based on a neural network with a quantile head paired with copulas to model dependency across the forecast horizon.