Streaming ML for Kotlin/KMP: online regression, bandits, sketches, calibration, and statistics.
-
Updated
Aug 17, 2026 - Kotlin
Streaming ML for Kotlin/KMP: online regression, bandits, sketches, calibration, and statistics.
Streaming Evaluation Protocol (Prequential + Resource Footprint + Calibration Drift)
Streaming fraud detection pipeline processing 1M+ transactions/day with sub-100ms latency and 97.3% AUC-ROC. Kafka + Spark Structured Streaming + XGBoost with a Feast feature store and Evidently AI drift monitoring.
Ultra-low latency (<20ms) polyglot streaming audio & ML pipeline featuring zero-copy POSIX shared memory, C11 AVX2/FMA SIMD DSP, Rust PyO3 runtime, Go Pion WebRTC gateway, and a real-time 60 FPS TypeScript HUD.
Add a description, image, and links to the streaming-ml topic page so that developers can more easily learn about it.
To associate your repository with the streaming-ml topic, visit your repo's landing page and select "manage topics."