SeafaringIT - MLOps for Energy Demand Forecasting | Hi Interns

Goals:
Build a production-ready ML system for energy demand forecasting
Optimize energy production, storage, and purchasing
Implement monitoring, retraining, and model improvement
Student roles: ML engineers, MLOps specialists, data engineers, backend engineers
Expected outcomes: MLOps pipeline with training, deployment, monitoring, and automated retraining for reliable predictions
Key features:
Dataset prep (consumption, production, weather) + feature engineering
Multiple time series models with comparison
MLflow for versioning/experiments
API deployment with batch jobs
Performance monitoring (latency, errors, concept drift)
Automated retraining
Operational guide, maintenance procedures
Cost and emissions impact analysis
Technologies: Python, time series libs (statsmodels, Prophet, LSTM), scikit-learn, MLflow, FastAPI, Docker, Docker Compose, CI/CD (GitHub Actions/GitLab CI), PostgreSQL, monitoring tools

MLOps for Energy Demand Forecasting