This planner helps you review a site, assess expected output, and understand the assumptions behind the result. The focus is on practical analysis, readable output, and clear source tracing.
The retrieval corpus gives the app a grounded explanation layer for solar fundamentals, weather impacts, costs, panel efficiency, and site-specific context.
NASA climatology supplies location averages, the ML model estimates irradiance, and the financial layer converts that estimate into savings and payback.
The LLM is not used to invent output. It explains the deterministic result and points back to retrieved documents when relevant.
This MVP is honest about missing data, seasonal variance, and the fact that estimates are not guarantees.
NASA climatology, monthly solar values, and saved browser state provide the current inputs.
Deterministic models compute the numbers, and the LLM explains them with grounded context.
Explanations show what was retrieved, what was calculated, and what remains uncertain.