I have an existing Python-based Serie A soccer prediction model and I'm looking for someone who can help me improve and automate it.
The model is already working, so I don't need someone to build a completely new system. The main work will be reviewing the current code, fixing some issues, improving the modelling and validation, and making the whole system more reliable.
Some of the initial work includes improving the hyperparameter tuning, standardizing metrics like Brier score and log loss, setting up proper walk-forward validation, and making sure there is no data leakage.
I also want to improve the existing Poisson/Dixon-Coles models, probability calibration, market-odds blending, EV calculations and betting thresholds.
Another important part is automation. The longer-term goal is for the system to collect data, update the database, generate predictions, compare them with the market, track results and CLV, run backtests, and test new models against the current production model.
I'm looking for someone who can work with the existing codebase and improve it step by step rather than rewriting everything unnecessarily.