Show HN: A Boring MLFlow Replacement

Category: infrastructure

Tags: mlops, experiment-tracking, model-registry, llmops, ai-platform

Score: 6.0/10 (Innovation: 5, Technical: 6, Documentation: 7, Utility: 6)

LUML is a platform for managing the complete ML lifecycle, offering experiment tracking, model registry, and deployment with a focus on data privacy by keeping storage and compute under user control. It differentiates itself with a unified AIOps framework that extends MLOps to LLMs and agents, and features client-side data transfers and pull-based execution models for enhanced security. Interesting for its privacy-first approach and modular architecture, though as a relatively new project it may lack the maturity and ecosystem of established tools like MLflow.

Target audience: ML engineers, data scientists, AI/ML teams managing the full model lifecycle from training to production

Repository: https://luml.ai/flow · Python · Apache-2.0 · 198 stars

View on Hacker News