EnerFlow blends your historical consumption with weather and occupancy data, then trains a Multiple Linear Regression model to forecast kWh usage, building by building, day by day.
Forecast · Next 12 days
Faculty of Engineering
0.94
Avg. R² Score
12%
Energy Saved
3s
Model Training
24/7
Predictions
Platform
From raw meter readings to AI-generated recommendations, EnerFlow gives energy managers the tools to plan, save, and report.
Manage multiple organizations, each with several buildings, universities, hospitals, offices.
Super Admin, Energy Manager, Analyst, each with the right permissions.
Temperature and humidity shape every prediction, because HVAC is honest.
Multiple Linear Regression trained on your data with a click. R², MAE, RMSE included.
Every forecast is scored against reality so the model keeps getting sharper.
Row-level security, encrypted secrets, no service role keys in the client.
Dashboard
Live cards for today's consumption, tomorrow's forecast, monthly totals and prediction accuracy. Six charts. Zero spreadsheets.
482 kWh
512 kWh
14.2 MWh
94.1%
Explainable, fast, deterministic. Trained in your browser in seconds.
AI Engine
We use Multiple Linear Regression for predictions (fast, explainable, academically grounded) and layer OpenAI on top for natural-language insights, recommendations, and a built-in energy assistant.
Train in seconds
80/20 split, normal equation.
Explainable
See every coefficient.
AI insights
Plain-English recommendations.
Chat assistant
Ask questions about your data.
Export PDFs, CSVs and Excel files with charts, predictions and an AI-generated executive summary — ready for the next board meeting.
Pricing
Pro
$49/mo
Unlimited buildings, AI insights.
Sign up in 30 seconds. Add your first building. Train your first model.
Start free — no card required