What problem does it solve? Automotive engineers and data scientists need to detect battery faults, sensor failures, and risky driving patterns from large volumes of vehicle telemetry, but building and deploying these ML models from scratch is time-consuming and error-prone. ## Core Features & Use Cases - Anomaly Detection: Isolation Forest and LSTM autoencoder implementations for battery, sensor, drivetrain, and charging anomaly detection with feature engineering pipelines. - Driver Behavior Analysis: Extracts acceleration, braking, cornering, speeding, and time-to-collision features from trip telemetry to compute safety scores for insurance and fleet management. - Production Deployment Guidance: Covers edge deployment (ONNX, INT8 quantization, Jetson targets), cloud fleet architectures (Kafka, Spark, MLflow), evaluation metrics, and monitoring checklists. - Use Case: Train an Isolation Forest on normal battery telemetry, deploy it to flag cell voltage imbalances and thermal anomalies in a fleet, and explain each flagged event by feature z-scores. ## Quick Start Ask the agent to build an Isolation Forest anomaly detector for battery telemetry data with feature engineering and model persistence.
