FAQ
Common questions about Rassket — what it does, how to use it, and what to expect from your results.
Do I need a data science team to use Rassket?
No. Rassket is built for energy analysts and operations teams. You connect your data, Rassket does the rest — no coding or technical expertise required.
What energy use cases does Rassket support?
Rassket supports five energy-specific domains: day-ahead load forecasting, solar and wind generation forecasting, energy theft and meter fraud detection, predictive maintenance for turbines, and short-term electricity price forecasting.
How long does training take?
Most models complete training in 15 to 40 minutes depending on dataset size. You can watch live progress throughout.
Is my data secure?
Yes. All data is encrypted at rest and in transit. Rassket is GDPR aligned and SOC 2 Type II certification is in progress. Enterprise customers can request on-premise deployment.
Can I export my results?
Yes. Results are exportable as PDF, CSV, or via API. Every export includes the forecast, confidence intervals, and feature importance summary.
What databases can I connect?
Rassket connects to PostgreSQL, InfluxDB, MySQL, MongoDB, and Prometheus. You can also upload files directly in CSV, Excel, Parquet, or JSON format.
What is the utility company context selector?
When you select your utility — such as TVA, EPB, or MTE — Rassket applies that organisation's operational context to your analysis, chat responses, and decision simulations. This makes results more relevant to your specific grid and customer base.
What happens if my model fails during training?
Rassket's self-healing engine automatically detects the error, diagnoses the cause, applies a targeted fix, and retries — without interrupting other models running at the same time. You will never see a raw error message. If a model cannot be recovered, you will receive a plain English explanation and a recommended next step.
Still have questions?
Review the full documentation or open Rassket to explore the product directly.