Find which script, notebook, data, command, or AI agent produced a file — locally, with evidence and honest uncertainty.
Research-first machine learning experiment tracker for comparing model metrics, scalar curves, artifacts, and experiment lineage.
Run Observation & Artifact Registration
This project implements a telco churn prediction pipeline with local Apache Airflow orchestration, native Apache Kafka streaming, and PySpark-based model training and inference.
Kimi Code / Claude Code / Codex 一体化桌面工作台：同一 GUI 无缝使用，多格式 AI 产物 Viewer、Diff 与时间机器
This project is a PySpark-based customer churn prediction pipeline that covers the full machine learning lifecycle, from raw data ingestion and preprocessing through model training, evaluation, and streaming inference. It uses MLflow for experiment tracking and artifact management, with a production-focused setup for reproducibility and monitoring.
Local-first governance and operational accountability system for AI work, with lifecycle structure, artifact tracking, and durable memory.
Crash Override — independent third-party profile of a public API surface, by API Evangelist. Crash Override is a software supply chain security company building a control plane for software development in the agentic era. It monitors code from developers and AI coding agents, inspects builds across CI/CD systems (GitHub Actions, GitLab CI, CircleCI
To associate your repository with the artifact-tracking topic, visit your repo's landing page and select "manage topics."