# Load Review

> Review interval electricity use against a building operating schedule and export an investigation record.

This is a personal software project by Prateek Mulye. It is free to use without an account. It does not represent an employer deployment or customer adoption.

## Workspace and source

- [Open Load Review](https://energy.prateekmulye.dev/)
- [Source code](https://github.com/prateekmulye/load-review)
- [Prateek Mulye's portfolio](https://prateekmulye.dev/)

## What it does

Inputs: One timestamp,kwh CSV, up to 20 MiB and 40,000 rows; confirmed timezone, interval and occupied schedule; optional timestamp,temperature_c CSV.

AI contribution: A small per-building TensorFlow.js model compared with weekday/time median and calendar regression baselines using chronological held-out readings.

Outputs: Findings CSV, original interval evidence CSV and HTML investigation brief.

## Review and limitations

One building meter. Model-based flags are withheld when the learned model does not beat eligible baselines. On three real building files evaluated, the model did not beat those baselines. No claim of equipment diagnosis, verified savings or general anomaly accuracy.

A person reviews the findings. The application does not treat model output as an authorized external action. Functional checks and synthetic examples do not establish operational accuracy or business impact.

## Data handling

Input contents, decisions and exports are processed in the browser. The app does not upload input contents or persist review sessions across reloads. Public runtime and model files are downloaded only when the user starts AI analysis; hosting providers can receive ordinary request metadata. Public model assets may remain cached. No account, analytics or input-content telemetry is required.

## Attribution

Author: Prateek Mulye, Senior Software Engineer and applied AI practitioner.